The geography of publishing in the Anthropocene
Abstract
One key aspect of the Anthropocene is the inherent disparities between the Global North and the Global South. These differences manifest in the causes and impacts of pollution, climate change, and species extinctions, but are they also present in the ways we write about the Anthropocene? We examine 77 peer-reviewed papers spanning 2009–2019 that explicitly feature conservation in an Anthropocene context. We compare these papers to a control group of papers that feature conservation but do not engage specifically with the Anthropocene literature. We found that both “Anthropocene” and “conservation” papers include a disproportionately large number of authors with affiliations in the Global North, despite half of the research taking place in the Global South. Moreover, this overrepresentation occurs regardless of author position or journal impact factor. We find that 84% of Anthropocene articles and 91% of conservation articles occurring in the Global North had a first author from the country of study, as opposed to only 55% of Anthropocene articles and 62% of conservation articles from the Global South. Studies occurring in the Global North almost always had at least one coauthor from the country of study (96% of Anthropocene articles and 97% of conservation articles). In contrast, only 81% of Anthropocene articles and 83% of conservation articles occurring in the Global South had any local coauthors. We used two text-mining algorithms to characterize the authorship networks and topics occurring in Anthropocene and conservation papers. These analyses showed that while both groups are interdisciplinary, Anthropocene papers had more distributed authorship networks and greater linkages across topics, and therefore have a flatter “topic surface” than the conservation papers. Our work suggests that conservation research programs that are explicitly grounded in the Anthropocene as a theoretical framework are more likely to reach across disciplinary lines.
1 INTRODUCTION
The term “Anthropocene” was introduced by Crutzen and Stoermer (2000) to characterize a new geologic epoch defined by humans' impact on the geologic history of the planet. The Anthropocene is a powerful way of thinking about humanity's impacts on ecological systems and has become increasingly popular in the scientific literature, moving beyond its geologic origin to be used in ecology and conservation biology, the humanities, and as a political term, among other fields (Corlett, 2015; Moore, 2016; Palsson et al., 2013). However, with the dawn of a new geologic epoch comes the realization of an unequal divide between those contributing to anthropogenic changes and those experiencing them, following a sharp Global North/Global South divide (Uddin, 2017). For example, the majority of CO2 impacts are produced by just three countries (China at 29%, the United States at 16%, and India at 7%—UCS, 2019); however, the impacts—coastal flooding, desertification, and civil unrest (Barnett & Adger, 2007) are felt keenly by individuals from countries with less resilient economies and government policies (UCS, 2019).
The Global South, particularly the tropics, is home to many biodiversity hotspots and presents opportunities for large-scale conservation (Brooks et al., 2006). Despite their high level of biodiversity, tropical regions are underrepresented in many aspects of ecological research including reviews of species' responses to climate change (Feeley, Stroud, & Perez, 2017), protected areas research (Blanco et al., 2020), and in biodiversity/ecosystem function (Clarke, York, Rasheed, & Northfield, 2017). Even within the tropics, there are publishing biases, for example, research locations in coral reef studies are positively correlated not with diversity, but with per capita income (Fisher et al., 2011). Systematic underrepresentation in the Global South may lead to these regions being under-considered in policymaking.
Thus, there is an inherent inequality to the Anthropocene, one that manifests itself not only in the scale and location of observable processes but also potentially in the scale and location of individuals who conduct those observations—scientists who study the Anthropocene. This overall disconnect between the Global North and South has been highlighted in ecological research and authorship. Stocks, Seales, Paniagua, Maehr, and Bruna (2008) found that 62% of the tropical studies in their survey had a lead author from another country, and 39% of multiauthored papers had only foreign scientists. Only 28% of the studies were international collaborations between local and foreign authors. Braker (2000) found that 75% of papers published in Biotropica in 1989 had local coauthors, but only 49% did in 1999. Similarly, in the field of biogeography over a 5-year period, 76% of first authorships from three leading journals were from Europe, North America, Australia, and New Zealand (Eichhorn, 2019). These inequalities in publishing also manifest themselves at the editorial level. A review of 20 leading conservation journals found that only 12.7% of editors were affiliated with institutions in Africa, Asia, and Central and South America and that several countries in South American and tropical Asia with very high biodiversity indexes were not represented by a single editor (Campos-Arceiz, Primack, Miller-Rushing, & Maron, 2018).
Because the Anthropocene as a topic of study is inherently multidisciplinary and far-reaching we wished to explore if published papers that examine conservation through the lens of the Anthropocene recapitulated the patterns of inequality that are present in other areas of the Anthropocene. Alternatively, because of the global nature of the Anthropocene projects we may expect research investigations using the Anthropocene as a theoretical lens would incorporate a diversity of methodologies and voices within their purview. As Inkpen and DesRoches (2019) say “The pervasiveness of the human footprint implied by the Anthropocene means that science today also requires mechanisms that allow natural scientists to engage more frequently with their colleagues in the social sciences and scholars in the humanities.” Given recent critiques of the knowledge deficit method of conservation (Toomey, Knight, & Barlow, 2017), there are calls for a shift to coproduction of knowledge and in doing so providing a greater diversity in conservation (Polk, 2015; Salomon et al., 2018). Because subsequent calls for transnational and transdisciplinary solutions are widely recognized in theoretical Anthropocene papers (e.g., Brondizio et al., 2016; Di Chiro, 2017; Ogden et al., 2013), we might expect to see an extension from theoretical to empirical research with a result of a more inclusive and global perspective on research in field Anthropocene papers.
In this paper, we examine these publishing trends in conservation papers that expressly articulate research about the Anthropocene. Specifically, we wish to: (1) Examine differences in local author involvement—lead author, last author, or any author—between field studies conducted in regions defined by predetermined Global North/Global South divisions. (2) Distinguish patterns of publication, prestige, and bias across Anthropocene research by examining patterns of content across journals and authorship across journal impact factors. Finally, we wish to (3) determine whether these variables differed between articles focused specifically on the Anthropocene and more general conservation biology work. We investigate these goals while simultaneously re-examining the conservation biology literature as a control, and in doing so also advance the broader conversation about inequalities in publishing in those realms by analyzing works decades after the seminal works of Braker (2000) and Stocks et al. (2008). Our research aims to examine these trends in publication bias surrounding the Global North/Global South divide in Anthropocene literature in light of the recent focus on conservation in a new era of biodiversity protection and coupled socioecological approaches toward conservation.
2 METHODS
2.1 Data collection
We searched the Web of Science Core Collection (Science Citation Index to 1985, Social Science Citation Index to 1985, Arts and Humanities Citation Index to 1985, Emerging Sources Citation Index to 2015) using the topics “Anthropocene” and “conservation” and restricted the results to papers published in or between 2009 and 2019. From this, we obtained 419 results which we narrowed to remove book reviews, review papers, and other results outside the scope of our study; we focused on papers with three criteria (a) site-specific studies with (b) a field component and (c) published in peer review journals. These studies included research in which fieldwork was conducted by the authors or compiled from sources such as regional climate data, previous field surveys, and citizen science databases. This yielded a sample size of 77 published papers about the Anthropocene. To develop a control data set, we paired each Anthropocene paper, with a randomly selected conservation-focused, but not Anthropocene-focused, paper from the same journal and same year. Three papers were published in the journal Anthropocene from which no controls were able to be obtained, therefore we compiled 74 “conservation” papers as controls. We also used the Web of Science to obtain data on the usage of both terms over time by searching for articles including “Anthropocene” and those including “conservation” and “Anthropocene.”
For each paper, we obtained the following information: journal title, impact factor, paper title, year, volume, pages, research location (city/state/country), first author name and location (city/state/country), last author name and location (city/state/country), and whether or not the first/last author was from the country where the research was taking place. We determined the author's location based on their author affiliation listed in the paper. If multiple affiliations were listed for an author, we took the first one, generally considering university affiliation to be a better indicator of permanent location than field station affiliation when both were listed unless outside information about the author indicated that the field station was their primary place of work.
We used institutional affiliation as a proxy for researcher location of origin (Braun & Bujdosó, 1983; Espin et al., 2017; Smith, Weinberger, Bruna, & Allesina, 2014; Stocks et al., 2008). We recognize that this may skew our results as, for example, a researcher from the Global South who had a primary affiliation at an institution in the Global North would count as a member of the latter in our analyses. As this bias works in both ways we hope that it has not systematically influenced our results—and future qualitative interviews could be conducted to further explore this phenomena.
For the purposes of this research, we defined the Global North as the United States, Canada, Western Europe, Australia, and New Zealand. We defined the Global South as all other countries (Independent Commission on International Development, 1980; Uddin, 2017). Locations of territories or commonwealths were treated as their own entities and not that of the governing country (i.e., The Macaronesian Islands were recorded as such and not as part of Spain/Portugal; Stocks et al., 2008).
To reduce reader bias in the scoring of Anthropocene articles to be used in this study, we selected a random sample of 10 articles from each individual reader's section. We then all individually read and approved articles for use in this study. The individual lists were compared and found to be identical. If any discrepancy had been found we would have discussed and come to a unanimous decision.
2.2 Textual analysis
To assess if Anthropocene and conservation articles represented distinct groupings focusing on different topics, or if those topics were linked through different epistemologies, we conducted a textual analysis of both Anthropocene and conservation biology articles using VOSviewer 1.6.13 (van Eck & Waltman, 2009; van Eck & Waltman, 2019). This tool generates co-occurrence networks of terms used in the scientific literature—in our case, terms from the titles and abstracts of the articles we examined. We used this to generate co-occurrence lists for Anthropocene and conservation biology lists. We also used co-occurrence data to create visual geographies of terms to further explore connections among and between closely networked terms (Bhattacharya & Basu, 1998; Lee, 2008). This analysis helps to visualize whether Anthropocene-focused articles are conceptualized differently within the broader conservation field (Bhattacharya & Basu, 1998; Lee, 2008).
2.3 Statistical analysis
To determine the difference in the proportions of authors located in the area of study between the Global North vs. the Global South, we used a two-proportion z-test. This process was repeated for first, last, and any author as well as for both the Anthropocene and conservation articles. We also used a two-sample t-test to determine the difference in average impact factor and number of citations/year between Global North and Global South articles for both Anthropocene and conservation biology categories. Lastly, we used a two-sample t-test to see if the average number of authors on a paper differed between Anthropocene and conservation articles. To determine if there were patterns in the perceived prestige of publishing across certain areas, we examined patterns of authorship as a function of impact factor. Specifically, we used simple linear regression to test whether papers with a higher proportion of authors from the Global South were more likely to publish in lower impact factor journals.
2.4 Geographic analysis
Finally, to investigate these patterns of coauthorship in a geospatial framework, we mapped patterns of collaboration using the R package “refsplitr” (Fournier, Boone, Stevens, & Bruna, 2020). Using the output from our Web of Science search, “refsplitr” disambiguates author names, parses author addresses, and georeferences author institutions. We then used the “plot_net_address” function to plot the coauthorship network based on georeferenced institutional addresses for each author.
3 RESULTS
Throughout our search timeline, the use of “Anthropocene” in scientific literature has exponentially increased. Of those mentions, only a small portion were conservation-focused articles as the number of articles meeting the “Anthropocene & conservation” search criteria increased only gradually (Figure 1). In both cases, 2013 was the year where the acceleration in the use of the term “Anthropocene” began.
We gathered 77 articles that met our search criteria for Anthropocene literature. Of these, 45 (58.4%) conducted research in field locations in the Global North and 31 (40.3%) in the Global South. Of the 74 articles meeting our search criteria for conservation journals, 32 (43.2%) conducted research in field locations in the Global North, and 42 (56.8%) in the Global South. The majority of research in the Global South was being conducted in Central and South America. The conservation journals showed a similar pattern but with more representation in African countries (Table 1). There were no significant differences in the geographic areas where studies were taking place between Anthropocene and conservation biology papers (X2 [DF 8, N = 73] = 1.6124, p = .80656). The list of 77 articles can be accessed and reused at https://digitalcommons.esf.edu/cgi/preview.cgi?article=1008&context=efb_fac.
| Anthropocene | Conservation | |
|---|---|---|
| Central and South America | 13 | 13 |
| Pacific Islands | 3 | 5 |
| Asia | 7 | 8 |
| Africa | 6 | 12 |
| Middle East | 0 | 1 |
| Other | 2 | 3 |
| Total | 31 | 42 |
The articles were published in a variety of journals; Anthropocene articles were generated from 54 different journals with the Global North publishing in 47 different journals and the Global South publishing in 30 different journals. Overall, the distribution of Anthropocene articles was skewed with the top four journals producing 22% of the studies and 90% of the studies being published in 46 journals (Supporting Information Figure S1).
In order to explore if Anthropocene and conservation focused articles dealt with different key terms or organized themes along different topologies, we used VOSviewer to mine the titles and abstracts of the selected articles for the co-occurrence of terms for our two sets of articles. We set our threshold to five usages of the term, which initially generated 66 terms for the Anthropocene articles and 42 terms for the conservation articles. We reduced this by removing irrelevant terms such as “hypothesis”, “data”, and "i.e." that were commonly used but not directly relevant to the content of the papers, and “Anthropocene” and “conservation” because they were the original search terms (Bhattacharya & Basu, 1998; Lee, 2008). We generated a list of terms and linkages for each data set (Tables 2 and 3). We also generated term density maps using 54 terms for the Anthropocene articles and 35 terms for the conservation biology articles (Figure 2a,b). The two sets of articles showed different patterns of term usages and linkages.
| Cluster 1 | Cluster 2 | Cluster 3 |
|---|---|---|
| Animal | Ability | Climate |
| Case study | Approach | Climate change |
| Consequence | Area | Decade |
| Decline | Biodiversity | Distribution |
| Diversity | Change | Environment |
| Extinction | Combination | Future |
| Forest | Ecosystem | Habitat |
| Importance | Ecosystem service | Impact |
| Loss | Factor | Individual |
| Plant species | Human | Majority |
| Population | Human activity | Model |
| Process | Increase | Period |
| Proxy | Land use | Presence |
| Range | Landscape | Response |
| Richness | Management | Site |
| Seed dispersal | Protected area | Time |
| Species | Region | Year |
| Species richness | Value | |
| Taxa |
| Cluster 1 | Cluster 2 | Cluster 3 | Cluster 4 |
|---|---|---|---|
| Biodiversity | Area | Change | Implication |
| Distribution | Extinction | Ecosystem | Landscape |
| Diversity | Interaction | Growth | Management |
| Elevation | Plant | Increase | Population |
| Forest | Presence | Model | Rate |
| Habitat | Process | Region | |
| Impact | Site | Relationship | |
| Importance | Species | Risk | |
| Pattern | Threat | Time | |
| Precipitation | Year | ||
| Temperature |
The Anthropocene articles had 1,177 links between terms and a total link strength of 3,459 (Table 2, Supporting Information Figure S2a). In this context, “links” represent co-occurrences between two keywords while strength of that link is a positive integer related to the frequency of those terms co-occurring across the literature. The sum of the strength indicates the interconnectivity within clusters (van Eck & Waltman, 2019). The higher this number the flatter the topic surface is, while low links can indicate “hills” within the topic topology resulting from insular key word clusters. The terms were grouped into three clusters by VOSviewer's “Network Visualization” function. Cluster 1 contained 19 terms, with “species” as the most connected one (53 links, total link strength of 430). As a whole, the cluster considered the study of species and populations. Cluster 2 contained 18 terms, with “change” as the most significant one (52 links, total link strength of 289). The cluster dealt with changes and conservation at the larger ecosystem and landscape-level scales. Cluster 3, the smallest of the clusters, contained “impact” as the most connected term with 52 links and a total link strength of 191. This cluster dealt primarily with predicting future changes.
The conservation biology articles had 506 links, for a total link strength of 1,645 (Table 3, Supporting Information Figure S2b). These terms were grouped into four distinct clusters by VOSviewer's “Network Visualization” function. Cluster 1 had 11 terms with “habitat” as its most-connected term, with 33 links and a total link strength of 146. Cluster 2 had 10 terms. “Species,” the most connected term, had 34 links and a total link strength of 250. Cluster 3 had 9 terms, with “region” as the most connected (33 links and a total link strength of 160). Cluster 4, the smallest grouping, contained only five terms, with “population” as the most connected 33 links and a total link strength of 126.
Unequal representation in authorship was apparent throughout both Anthropocene and conservation biology articles. Studies conducted in regions of the Global South were less likely to have a first (two-proportion z-test, p = .008) or last (two-proportion z-test, p < .001) author based in the country of the study compared to research conducted in the Global North (Table 4). The same representation pattern occurred in conservation biology articles (two-proportion z-test: First Author p = .002; Last Author p < .001; Table 4). Additionally, only 80.6% of articles conducted in Global South regions had any author based in the region on the paper which was significantly less than Global North regions with 95.5% of articles having an author based in the country of study (two-proportion z-test, p = .054). The same disparity occurred in conservation biology articles with 83.3% and 96.9% of papers having an author based in the region of study in Global South and Global North, respectively (two-proportion z-test, p = .038).
| Anthropocene | Conservation | |||
|---|---|---|---|---|
| Global North (n = 45) | Global South (n = 31) | Global North (n = 32) | Global South (n = 42) | |
| First author | 0.844 | 0.548 | 0.906 | 0.619 |
| p = .008 | p = .002 | |||
| Last author | 0.867 | 0.467 | 0.875 | 0.524 |
| p = .000 | p = .000 | |||
| Any author | 0.956 | 0.806 | 0.969 | 0.833 |
| p = .054 | p = .038 | |||
- Notes: p-Values indicate the significance of the difference in proportion between the Global North and the Global South for each author category.
There was no difference in the number of citations/year between the Global North and Global South within Anthropocene articles (p = .227; GN: x¯ = 2.50, s = 3.58; GS: x¯ = 1.72, s = 1.97). We explored two factors which may influence citation, impact factor as a proxy for local vs. international focused journals (Stocks et al., 2008) and number of coauthors (Abt, 2017). There was no difference in the mean impact factor between Anthropocene and conservation articles nor the Global North/Global South (Supporting Information Table S3). When exploring the relationship between authorship and impact factor, we found no relationship between the proportion of authors based in the Global South and the impact factor in which the article was published (Figure 3; Anthropocene: R2 = 0.01, conservation: R2 = 0.12). Lastly, there was no difference in the mean number of coauthors per paper between Anthropocene and conservation articles (p = .277; Anthropocene: x¯ = 5.18, s = 3.37; conservation: x¯ = 5.82, s = 3.85).
The coauthorship maps produced using the R package “refsplitr” showed different patterns of interconnections between Anthropocene and conservation papers (Figure 3). The conservation map shows a concentration of networks between Global North countries, particularly the United States and Europe. The distribution of nodes (institutions) is widespread, but most networks in the Global South lead to nodes in the Global North in the conservation maps. The Anthropocene map shows a greater distribution and strength of networks throughout the Global South and establishes connections between Global South countries, particularly South East Asia to South America. The connection between the United States and Europe is much weaker in the Anthropocene map, and the map adds numerous Global North locations like Alaska and northern Canada. The number of nodes stayed relatively the same between maps.
4 DISCUSSION
Our analyses demonstrate that there is a Global North/South divide present in Anthropocene scientific literature and that those inequalities are mirrored in the broader conservation biology literature. We found that there was a strong disparity in local author involvement; local authors based in the Global South were frequently, and significantly, underrepresented in published literature, especially as the first or last author on a paper. By analyzing a large number of recent Anthropocene articles, our method was the first step in identifying global inequalities in Anthropocene related publications, and at the same time, we extend a line of inquiry following publishing inequities in the broader fields of ecology and conservation biology.
Through our co-occurrence of terms analyses, we demonstrated that the Anthropocene literature exists as a distinct area within conservation biology (Lee, 2008). This information provides support for our claim of the distinction around general conservation biology and Anthropocene research. Thus, our analysis of the difference in author representation and location is distinctive.
We saw an underrepresentation of local authors for both our Anthropocene and general conservation biology studies that took place in the Global South. Only 55% of Anthropocene articles and 62% of conservation articles with research from the Global South had local first authors, both of which are significantly lower than articles where research took place in the Global North (two-proportion z-test, p = .002 and p = .001). This trend is repeated for the last authors, with only 47% of Anthropocene articles and 52% of conservation articles containing a local author (p < .001 in comparison to Global North articles for both).
However, this authorship discrepancy was not as extreme as in previous reports on tropical ecology research. Stocks et al. (2008) found that only 38% of first authors were based in the country of study, as opposed to our 62% for conservation biology articles published between 2009 and 2019. This number was higher for overall authorship—83% of all conservation studies that we reviewed from the Global South had at least one author from the country of study. It is a promising sign that increasingly large proportions of studies are including at least one local author; however, a major gap still exists when it comes to the “prestige” positions of first and last author.
Many regions of the tropics are underrepresented in scientific research—Stocks et al. (2008) found that only 62% of tropical countries were represented in articles published in two leading tropical ecology journals between 1995 and 2004. Additionally, almost two-thirds of studies were conducted in just 10 countries, with one-quarter of the papers coming from Brazil and Costa Rica. This is not a new trend, or even a new concern; 74% of tropical research articles published in Ecology and Biotropica in 1983 and 1984 were from 10 countries (Clark, 1985). We found a similar trend in the literature that we reviewed—only eight articles meeting our criteria were published with research in the Pacific Islands between our Anthropocene and conservation biology groups, and research occurring across the continent of Africa was also underrepresented, especially in Anthropocene studies as compared to conservation biology papers.
We acknowledge that many studies published in local and national journals in the Global South are not included in the database. This lack of available research to the Web of Science is a major concern when it comes to setting conservation priorities since the full biodiversity of a region may not be taken into account and the region may end up under-prioritized, or the results may be biased due to an underrepresentation of tropical species (Brown et al., 2016; Feeley et al., 2017; Lenoir & Svenning, 2015).
Our analysis of the relationship between author geographic location and journal impact factor showed no strong relationships. Stocks et al. (2008) had hypothesized that one reason for lack of representation for authors from the tropics was that there were incentives for them to publish in regional journals that were not included in the original analysis. As these journals tend to be smaller and not specialized we would anticipate a skewed representation with articles in lower impact factor journals having a disproportionately high level of authors from the Global South. Our work, however, suggests that there are no clear trends—and that representation of authors from the Global South is equally poor across a wide range of journal impact factors.
Similarly, our analysis of average author count per paper showed no difference between Anthropocene and conservation articles. If we are assuming that papers published with an Anthropocene focus are more inclusive and cross-disciplinary, we might expect to see more authors on Anthropocene papers. Anthropocene articles would be incorporating more disciplines, which would be reflected in the number of authors on the paper. However, the breadth of disciplines in Anthropocene articles was not reflected in the number of authors on a paper.
The “Anthropocene” selection of articles showed different patterns of term usage and linkages compared to the more general conservation articles, showing its distinction as a field and suggesting future opportunities for interdisciplinary work and increasing collaborations. The Anthropocene articles had more overlap in their term usage and tighter links between term usage across papers—they had 19 more recurring terms than the conservation articles, and yet these terms were grouped into only three clusters as opposed to four clusters from the conservation articles (Tables 2 and 3). The Anthropocene articles had 1,177 linkages between the 54 terms and a total link strength 3,459, whereas the conservation articles had 506 links between 35 recurring terms for a total link strength of 1,645. These results are mirrored in our visualizations (Figure 2a,b), which show a similar topography of low-lying islands but different topology with different degrees of clustering. The main “islands” in the conservation biology papers are made up of “species” and “populations,” and most of the other terms are very spread out. The main islands in Anthropocene articles are made up of “species” “landscapes” and “change,” reflecting a broader, more processed, and place-based research focus for Anthropocene articles. There is also more clustering and linkage across different islands in the Anthropocene visualization resulting in a flatter topic surface.
The Anthropocene papers showed a greater distribution and interconnectedness of coauthorship networks throughout the Global South (Figure 3). While we have shown that local author inclusion is only about 80% for both conservation and Anthropocene articles, these maps show a clear increase in coauthorship connectedness in the Global South in Anthropocene articles compared to conservation articles. This suggests Anthropocene research is on the way to becoming more global and transdisciplinary. In the conservation articles, we observed fewer connections between Global South and Global North countries.
The field of Anthropocene conservation, while primed for reinforcing inequalities observed in other disciplines, is also an opportunity for improving research conduct. Conservation research in recent decades is filled with geographic biases; the locations where conservation research is needed most are the same locations least represented in the literature (Wilson et al., 2016). Highest publication densities come from areas with high affluence, high biodiversity, or both (Hickisch et al., 2019; Meijaard, Cardillo, Meijaard, & Possingham, 2015). And at the same time, local ecological knowledge is often underutilized (Brook & McLachlan, 2008) despite offering potentially powerful insights into conservation and restoration (Kurashima, Jeremiah, & Ticktin, 2017). Our analysis of Anthropocene literature shows the same biases as conservation biology journals (Table 4). Yet, in order to tackle the largest problems faced by this planet—issues deeply rooted in the causation of the new geologic epoch—local, indigenous aid, and knowledge will be critical (Brook & McLachlan, 2008). The field of conservation biology, a field expanding to include the “Anthropocene,” is bringing the inequalities it possesses into this new geologic epoch.
4.1 Conclusions: Doing conservation in the Anthropocene
The Anthropocene, as a geological epoch, is an inherently spatiotemporally rooted concept—we understand it as a place in time and space, and therefore it makes sense that “change” and “landscapes” would be used frequently in Anthropocene publications. This underscores the differences between Anthropocene focused conservation and more general conservation biology publications, yet the overall flat topographies within and between maps (Figure 2a,b) suggests that the field is broad and there are few ideological hills upon which researchers are dying. Moreover, when we look at the landscapes of collaborations, we see a more diffuse network of collaborations within Anthropocene papers, with greater prominence of nodes in the Global South (Figure 3a,b). In these ways, we suggest that the geography of publishing in the Anthropocene provides opportunities for multiple disciplines to contribute, perhaps befitting a term that has been widely adopted outside of its original, geologic, foundations.
Our results show some improvements from previous regional literature through the increasing inclusion of local authors from the Global South, a trend mirrored in the related field of biogeography (Ladle, Malhado, Correia, dos Santos, & Santos, 2015). Yet despite recent improvements, on a broader scale, our results show consistency with the inequalities still exist within Anthropocene field studies. Not only should more research into these disparities be conducted but research in how to decrease this disparity and increase Global South authorship should be investigated. These partnerships could provide far-reaching insights into important ecological processes that could shape the global response to the Anthropocene.
Both the Anthropocene and conservation biology literature displayed biases toward nonlocal authors, and further research should be conducted to determine if this trend is consistent for more general areas of research such as ecology. The disparity in publication indicates a lack of knowledge exchange that is important for science to address and understand global environmental issues. In doing so these inequalities perpetuate the neocolonial history of conservation biology research (Baker, Eichhorn, & Griffiths, 2019), and the erasures of historical violence, and indigenous labor, inherent to the Anthropocene (Yusoff, 2018). Thus, continued progress is necessary for improving the inclusion of voices from the Global South, especially as first and last authors.
How do these results help improve conservation and practice, though? While quantifying problems is an important first step, the existence of that information is not sufficient to effect change. While many conservation problems were initially seen within a knowledge-deficit framework, recent work has shown that translating conservation findings into actionable results requires interactions and discussions that recognize different epistemologies and power differences (Toomey, 2016). When solutions are the result of a process of coproduction of knowledge, we see more equitable and potentially successful outcomes (Nel et al., 2016; Norström et al., 2020). In the context of our work, we can see that Anthropocene papers tend to have more interconnected keywords, incorporate more authors from the Global South and have more distributed authorship networks. The Anthropocene as a theoretical framework, which includes investigating problems from a transdisciplinary perspective, and having more diverse groups of people involved within those perspectives, is suggestive of a knowledge coproduction framework which may ultimately portend greater conservation success in these outcomes.
ACKNOWLEDGMENTS
This paper arose as part of a class assignment for EFB 696 Anthropocene conservation and the authors would like to thank S. Howard and T. Yakovleva for their contributions to class discussions. We would also like to thank an E. Bruna and an anonymous reviewer for their kind and helpful reviews.
CONFLICT OF INTEREST
This was a self-funded research project and therefore we have no conflict of interests to declare.
AUTHOR CONTRIBUTIONS
Conceived the project: Megan A. Hazlett, Kate M. Henderson, Ilana F. Zeitzer, Joshua A. Drew. Collected data: Megan A. Hazlett, Kate M. Henderson, Ilana F. Zeitzer. Analyzed data: Megan A. Hazlett, Kate M. Henderson, Ilana F. Zeitzer. Wrote manuscript: Megan A. Hazlett, Kate M. Henderson, Ilana F. Zeitzer, Joshua A. Drew.
AUTHOR POSITION STATEMENT
Finally, we feel it is important to acknowledge that this paper, while investigating the disparity between publishing in the Global North/South, is published by authors only from the Global North. We carry biases from this position and in the spirit of transparency wish to make them explicit. We also would like to acknowledge with respect the Onondaga Nation, firekeepers of the Haudenosaunee, the indigenous peoples on whose ancestral lands SUNY ESF now stands.
ETHICS STATEMENT
This project was conducted in accordance with the Society for Conservation Biology's Code of Ethics. It did not involve human subjects or living organisms.
