Seeing the Forest for the Trees Sequel I: An Extension of the 1985–2017 Bibliometric Analysis of Environmental and Resource Sociology

Abstract Twenty years after the organized sessions on environmental sociology (ES) and natural resource sociology (NRS) at the 2000 International Symposium on Society and Resource Management, a featured collection on environmental and resource sociology came out in Society and Natural Resources. The four commentaries included in the special section provide insightful responses that help to clarify, strengthen, and expand the points we made in our bibliometric analysis article. Here, we present an extension of the previous analysis using more recent journal article collections (2017–2022), while incorporating responses to colleagues’ major comments on our original article. The new bibliometric analysis of environmental and resource sociology suggests increasing cross-linkages between the ES and NRS subfields. It would be meaningful to conduct similar analyses of non-English counterpart literature in future research. Further dialogues should also shift the focus to diverse perspectives, experience, and practices of individual researchers, particularly emerging ES/NRS scholars.

Environmental sociology; ES-NRS 2.0; extended bibliometric analysis; global context; increasing synergies; individual reactions; natural resource social science; natural resource sociology Two decades have passed since the organized sessions on Environmental Sociology and the Sociology of Natural Resources at the 2000 International Symposium on Society and Resource Management (ISSRM) and the release of the eponymous special issue (Vol. 15,No.3) of Society & Natural Resources (S&NR) in 2002 (Buttel and Field 2002).Since there are only a few research articles on this topic in recent years (e.g., Krannich et al. 2013;Qin and Flint 2010;Qin et al. 2018), the journal's 2020 featured collection on a bibliometric analysis of environmental and resource sociology (Vol.33, No. 9) provides another important forum for examining the relationships between environmental sociology (ES) and natural resources sociology (NRS).As the terms "environment" and "natural resources" often conjure up different conceptions of the natural world (Freudenburg 2002;Qin and Flint 2009), these two subdisciplines tend to represent somewhat distinct definitions of environment and scopes of research.ES mostly addresses metropolitan-based environmental pollution or degradation on relatively broad scales (e.g., nation-state or globe), whereas NRS typically focuses on the use and management of natural resources in nonmetropolitan settings and emphasizes a community or regional scale of analysis (Buttel and Field 2004).In our previous bibliometric study (Qin et al. 2020), we explored the coauthorship and citation networks in environmental and natural resource sociologies (as well as related social science disciplines).The results and visualization illustrated the coexistence of two different, but not isolated, research traditions and communities.This analysis further advanced the discussions on the differences and similarities between two main subdisciplines in environmental and natural resource social sciences.The insightful responses from colleagues representing both ES and NRS are highly constructive as they go beyond typical commentaries and help to clarify, strengthen, and expand many of the points we made in our article (Krannich 2020;Ulrich-Schad and Givens 2020;Zhang and Liang 2020;Zinda 2020).
As the only scholar who contributed to both S&NR collections on environmental and resource sociology, Krannich (2020) argued that we should now turn our attentions from the ES-NRS distinctions to the feasibilities, pathways, and strategies of ES-NRS synergies.This may be considered an evolution from ES-NRS 1.0 to ES-NRS 2.0, similar to the upgrade from Climate 1.0 to Climate 2.0 (the shift from the debate on anthropogenic climate change to the focus on mitigation of and adaptation to climate change impacts).A major theme of existing research on the interactions between ES and NRS consists of implications for the professional and career development of emerging researchers in relevant fields.Ulrich-Schad and Givens (2020) extended the dialogue between the two subdisciplines through a pilot study of the identities, experience, and perspectives of early-and mid-career environmental and natural resource sociologists.Stressing the multi-paradigmatic nature of environmentally oriented social science, Zinda (2020) also advocated for broadening intellectual realms and cultivating diverse niches for future generations of interdisciplinary ES/NRS scholars.Although our analysis and the first three commentaries had a clear focus on the United States context, Zhang and Liang (2020) continued this line of inquiry by exploring the applicability of bibliometrics to studying scholarly networks on societal-environmental interactions in China.We also appreciate their efforts to make connections to an earlier S&NR essay on developing a transdisciplinary Chinese environmental and resource sociology (Qin and Flint 2010).
In the former bibliometric analysis paper, we highlighted the needs for tracking the ES-NRS research trends through follow-up studies and for conducting comparative analyses in broader geographic settings.These ideas were generally endorsed by authors of the four thoughtful commentaries.We feel it is meaningful to enlarge both the temporal and spatial scopes of our previous analysis considering the rapidly growing ES/NRS literature and the groundwork for extending such work internationally.Since the initial study covered an extensive time range , its analysis and results mainly involved established and middle-career scholars in environmental and resource sociology (or environmental and natural resource social sciences).Here, we present an extension of the initial bibliometric analysis while incorporating replies to colleagues' relevant comments on the methods and findings of the original article.Our primary research objective is to examine how those coauthorship, citation, and keyword networks may change when using updated data that better represent younger scholars from ES/NRS-related areas of study.

Methods
The procedures of data collection and analysis for this research are generally consistent with those of the earlier study so as to ensure their results are comparable.In Qin et al. (2020), we adopted a combination of purposive sampling and keyword-based search processes to select 8027 articles published during 1985-2017 from a range of journal sources, including six journals of particular interest to ES and/or NRS scholars (e.g., Society & Natural Resources, Environment & Behavior), fifteen general sociological journals (e.g., Rural Sociology, Social Science Quarterly), and nine interdisciplinary natural resource science journals (e.g., Journal of Forestry, Journal of Soil & Water Conservation).Some potentially relevant journals were excluded from the data collection because of relatively short publication history (e.g., Environmental Sociology, Sociology of Development, and Energy Research & Social Science), none or limited inclusion in the Scopus database (e.g., Journal of World System Research), and our focus on the American academic setting (e.g., International Journal of Comparative Sociology, Land Use Policy).Several other journals with broad scopes, such as the Proceedings of the National Academy of Sciences, Nature, and Landscape & Urban Planning were also not included considering our emphasis on disciplinary backgrounds related to sociology and the social or human dimensions of natural resources.We also left out books because they were less suitable for our bibliometric analysis (particularly the coauthorship network analysis) than journal articles, and because we needed to keep the search processes manageable.Nevertheless, we agree with our colleagues that future studies could widen the selection of publication sources depending on specific research purposes and questions (e.g., to identify the most influential/productive scholars or seminal publications in ES and/or NRS).
The article dataset for the previous bibliometric analysis consisted of all review and research articles from those six target journals and additional articles selected from 24 sociological and natural resource journals using a series of keyword combinations. 1These keywords were developed following key research themes and areas in ES and NRS.Although the list did not encompass every single topic in the two subfields, we believe these terms (particularly those more general ones such as "environmental," "sustainable," and "social") could capture a wide range of problems and issues.We adjusted the article search process for this extended analysis as it involved a shorter study period (2017-2022) and a substantially less amount of material.We still included all research articles and reviews from several target journals especially related to ES/NRS.Because of recent changes in the publication outlets for ES research, we substituted Environmental Sociology for Organization & Environment in this stage.Next, instead of repeating the keyword-based search approach with the other sample journals, we identified relevant literature by manually checking the titles and/or abstracts of all articles recently published in these sources. 2 As there were too many records (3592) retrieved from Water Resources Research for 2017-2022, we replaced it with two other journals closely related to natural resource social science -Annual Review of Environment & Resources and Fisheries.The literature search for this follow-up study was completed in September 2022 and generated a bibliometric dataset including 3270 articles (see Table 1). 3 We replicated our earlier analyses of descriptive journal/author statistics and bibliometric networks using the assembled data.VOSviewer (Version 1.6.5)was again used to visualize coauthorship (scholar collaborations), citation (direct citation relations), bibliographic coupling (citations of common sources), and keyword co-occurrence (co-presence of author keywords) networks. 4In light of the comments made by Krannich (2020) and Zinda (2020), we think it is necessary to further elaborate on our strategies for classifying scholars included in the mapping of networks.The classification approaches were mainly guided by researchers' affiliations and institutions: ES scholars ¼ sociologists or social scientists in liberal arts sociology and related departments; NRS scholars ¼ sociologists or social scientists in agricultural social science or natural resource departments and natural resource agencies.However, we also consider other relevant factors such as academic backgrounds, self-identification, and professional connections.For example, some scholars situated in Sociology programs were nonetheless placed in the NRS category as they mainly engaged in the Natural Resources Research and Interest Group (NRRIG) of the Rural Sociological Society (RSS) and/or the International Association for Society and Natural Resources (IASNR).Although we made every effort to ensure our classification results were accurate and consistent, it was possible that a few scholars were allocated to subfields they did not truly identify with.Such limited discrepancies should have negligible effects on the general patterns of those bibliometric networks produced in VOSviewer.It is also worth noting that the locations of individual scholars in the network maps were determined by the collaboration and citation linkages among them instead of their assigned groups.For the most part, our analysis evaluated the alignment of coauthorship, citation, and bibliographic coupling relations with a key ES-NRS differentiating criterion based on scholars' academic homes (Buttel 2002;Qin et al. 2018).Whether the scholar groupings match individual researchers' own intellectual identities is another empirical question that needs to be addressed in future research.

Results
Journals with the most numbers of articles in the new dataset were (numbers of articles in parentheses): Environmental Management (530), Society & Natural Resources (528), Human Ecology (383), Environment & Behavior (239), and Environmental Sociology (215).Altogether, these five journals accounted for 58.0% of the 3270 articles included in the analysis.The annual number of published articles was quite stable during the 2017-2022 study period, with an average of 565 articles for each year (see Figure S1 in the Supplemental Online Material). 5While this suggests an increased rate of literature growth in ES and NRS, the relatively higher number of publications per year in the extended analysis might also result from those journal replacements and the adjusted literature search approach.Among the 9500 authors included in the analysis, 442 had three or more publications in the article database.About 23.5% of them (104) were in liberal arts sociology and associated programs, while the rest (76.5%) were from natural resource-related departments or organizations.Major countries or areas represented in the author affiliations include (numbers of articles in parentheses): United States (1990), Canada (265), United Kingdom (265), Australia (206), Germany (152), Sweden (120), Netherlands (115), mainland China (109), Brazil (73), and Spain (73).
Figure 1 shows the new coauthorship networks consisting of all 442 authors with three or more articles in the database.We chose this selection criterion for scholars considering the shortened time period for this study and the large majority of natural resource sociologists/social scientists in our dataset.In other words, we tried to ensure enough ES and NRS scholars for the analysis by including more authors.As shown in the figure, many individual scholars or small groups of scholars were not linked to the central, connected section, where ES and NRS researchers were generally mingled together (albeit dominated by natural resource sociologists or social scientists).Several sets of authors in the peripheral areas also indicated collaborations across different categories of scholars.
The updated scholar citation and bibliographic coupling networks are presented in Figure 2 and Figure 3, respectively (see Figure S2 in the Supplemental Online Material for a zoomed-in look of the core section of Figure 3).The scholar citation networks show the largest set of connected authors (357) in the analysis, while the coupling networks covered all 442 authors meeting the threshold of publications (three or more articles).Overall, although researchers with similar disciplinary backgrounds still tend to be located close to each other in these two maps, the ES-NRS clustering here is much less clear compared to the original results and there were many more overlaps between the two groups of scholars.
Unsurprisingly, the analysis of journal citation and bibliographic coupling networks again produced two clusters of journals (see Figures S3 and S4    Resources held important roles in linking the two journal groups.Moreover, more than half (56.6%) of the records in the author keyword co-occurrence networks were also obtained in the previous (1985-2017) bibliometric analysis.Whereas there was no clear pattern of keyword co-occurrence relations in the earlier study, two of the three keyword clusters identified in this analysis (see Figure 4) match well with the focal areas of ES and NRS, respectively (e.g., "environment" and "environmental justice" for ES; and "natural resources" and "social ecological systems" for NRS).They are also connected by a middle field revolving around terms related to climate change (e.g., "vulnerability" and "resilience").Those keywords with the highest occurrences (e.g., "climate change," "conservation," "sustainability," and "environmental justice") were situated near the centers of their respective clusters.

Discussion
In this study, we extended our 1985-2017 bibliometric analysis of environmental and resource sociology using more recent research articles from selected journals.The relatively brief period for this research should be sufficient since our primary purpose is to identify changing trends in the ES-NRS interactions.In contrast with results of the original analysis, the new scholar network patterns based on the latest literature demonstrate more cross-linkages between the two research traditions.Despite the adjustments in journal selections and article search processes, we believe the expanded analysis reflected actual evolvement in relevant fields of research as our data collection and analytical approaches remained mostly consistent.Whereas our previous bibliometric study focused on a "forest-level" view of environmental and resource sociology, this extended analysis targeted the "forest regeneration" process in ES and NRS.Following the approach adopted in Janssen (2007), we also reran these analytic procedures using the whole 1985-2022 article dataset (10945 articles in total).The results largely echoed those of the 1985-2017 bibliometric study since the aggregate analyses still emphasized late/middle-career rather than early-career scholars.In their thought-provoking commentary, Ulrich-Schad and Givens (2020) shared a "forest in the future" perspective focusing on emerging ES/NRS scholars.The updated author network analyses encompassed significantly more scholars because of a lower selection criterion (м3 articles vs. м6/10 articles previously).Most authors (87.1%) included in the results of follow-up analyses were not part of the earlier scholar networks. 6Overall, these researchers represented relatively younger cohorts of practitioners, while many of them were at an early or middle stage of their professional careers.The collaboration and citation relations among these scholars and between them and established researchers all contributed to changes in the newly generated bibliometric networks.The less distinctive ES-NRS division shown in this analysis is also in agreement with the results of a recent study of the evolving central tendencies within the two subfields (Qin et al. 2018).
The dots in the author network maps can be seen as various types of members in ES and NRS (e.g., "cadre," "bridgers," and "vanguards") instead of specific scholars.As existing discussions on the ES-NRS relationships have remained at rather general scales, it is still not clear how individual researchers view the interactions of these two subfields and how they identify their own specialties.The exploratory analysis in Ulrich-Schad and Givens (2020) laid solid groundwork for further research in this direction.It is also our personal observation that American graduate student training in ES has largely overlooked the NRS tradition and thus many ES scholars may not be aware of the ES-NRS distinctions and similarities.This is probably even more true in non-U.S.academic settings.Therefore, it would be meaningful to carry out similar bibliometric analyses of counterpart literature in other languages such as Chinese and Spanish.
Considering the interdisciplinary nature of our larger field, the scope of the ES-NRS dialogue should also be expanded to engage other related social and natural science subdisciplines.We concur with Zinda (2020) that it would be very informative to continue the "forest in the future" initiative with a brief survey of members of relevant domestic and international professional societies, including the NRRIG of the RSS, the IASNR, the ES sections of the American Sociological Association and the Chinese Sociological Association, Research Committee 24 (Environment and Society) of the International Sociological Association, and the Social Science Working Group of the Society for Conservation Biology.Such efforts will allow us to not only see the forest for the trees, but also see the trees for the forest.

Concluding Remarks
If knowledge is power, then knowledge about knowledge (meta-knowledge) is like a mirror that scholars can use to reflect on their scientific identities, paradigms, and practices.We genuinely hope that our bibliometric analyses can help the ES-NRS discourse move beyond a simplistic debate to explore their constructive synergies.Whereas the growing cross-linkages shown in the updates are encouraging, there is no standard formula for developing such convergences in future research.Previous studies have suggested concrete and creative ways of integrating the expertise from both subfields in a range of promising areas, such as the political economy of the environment and natural resources, environmental justice and environmental movements, the social construction of the environment and landscapes, and community-based environmental conservation or natural resource management (Belsky 2002;Buttel 2001;Buttel and Field 2004;Field, Luloff, and Krannich 2002).Our updated keyword co-occurrence network analysis again highlighted the potentials of climate change and sustainability studies for developing ES-NRS linkages.We have much to learn from those forerunners (e.g., Frederick H. Buttel and William R. Freudenburg) whose work well embraced the realms of both ES and NRS.Further analyses of individual scholars' perspectives and experience will also help to answer many of the insightful questions raised in Krannich (2020).Since the organized ES-NRS session at the 2000 ISSRM, an increasing number of researchers have associated themselves with both ES and NRS or self-identified as environmental and resource sociologists/social scientists.In addition to continuing investigations, ES-NRS 2.0 can go a long way by entailing more actions in response to the call for synthesizing the sociology(ies) of the environment and natural resources, in both the U.S. and global contexts.
Notes 1. See Table 1 in Qin et al. (2020) for a full list of these keyword combinations.2. The lead author and one of the coauthors quickly evaluated each of the articles collected in this stage.3.In total, the six target journals provided 1574 articles, while 1696 articles were selected from the 10046 records retrieved from the other 25 sociological or natural resource science journals.Despite the relatively short period of study, this dataset is over two-fifths (40.7%) the size of the one used for the earlier bibliometric research (8027 articles).4.More details on the analytical procedures can be found in the Methods section of Qin et al. (2020).Although Version 1.6.18 of the VOSviewer software (https://www.vosviewer.com/) was available at the time of data analysis, we chose to use the older version so that the results of the original and extended studies were comparable. 5.When calculating the average number of articles per year, we weighted the number for 2022 based on the actual time range covered in the data collection.6.The scholar coauthoship, citation, and bibliographic coupling networks in the original analysis included 178 authors in total.Among them, only 57 appeared again in the new bibliometric networks.
in the Supplemental Online Material).One primarily consisted of Environmental Sociology, Environment & Behavior, Population & Environment, and those general sociological journals.The other included Society & Natural Resources, Environmental Management, and natural resource

Figure 1 .
Figure 1.Scholar coauthorship networks (2017-2022).The figure includes all 442 authors with three or more article in the full dataset: red circles ¼ sociologists or social scientists in agricultural social science or natural resource departments and natural resource agencies, blue circles ¼ liberal arts sociologists or social scientists.The sizes of circles are weighted by the number of articles.The widths of lines denote the strength of coauthorship relations.

Figure 2 .
Figure 2. Scholar citation networks (2017-2022).The figure includes 357 (out of 442) connected authors with three or more article in the full dataset: red circles ¼ sociologists or social scientists in agricultural social science or natural resource departments and natural resource agencies, blue circles ¼ liberal arts sociologists or social scientists.The sizes of circles are weighted by the number of articles.The widths of lines denote the strength of citation relations.

Figure 3 .
Figure 3. Scholar bibliographic coupling networks (2017-2022).The figure includes all 442 authors with three or more article in the full dataset: red circles ¼ sociologists or social scientists in agricultural social science or natural resource departments and natural resource agencies, blue circles ¼ liberal arts sociologists or social scientists.The sizes of circles are weighted by the number of articles.The widths of lines denote the strength of bibliographic coupling relations.

Figure 4 .
Figure 4. Author keyword co-occurrence networks (2017-2022).The figure includes 53 keywords with 20 or more occurrences in the full database.Colors stand for different clusters generated in VOSviewer.The sizes of circles are weighted by the number of occurrences.The widths of lines denote the strength of co-occurrence relations.A color version of the figure can be found in the online publication.