3.1 Overall characteristic analysis
Through the descriptive statistical analysis of the data, the annual publication distribution and word frequency of China’s sustainable development research are drawn (Fig. 1). We also plotted an exponential curve matching annual publication volume and found significant overlap between the two. This proves that from 1991–2021, the annual publication volume of China’s sustainable development research showed exponential growth. This trend will continue for the foreseeable future as China pays more attention to sustainable development and the construction of its ecological civilization (J. Wu & Bai, 2022). Based on the key points of the annual publication volume, it takes about 10 years to go from 10 to 100, and then from 100 to 500. However, from 500 to 1000, and then from 1000 to 2000, this time is shortened to only about 2 years. This also confirms the trend of exponential growth from the side, which is especially obvious after 2015. Based on word frequency analysis, the word frequency trends of the two core keywords “Sustainable Development” and “Sustainability”, are extracted as shown in Part A and B in Fig. 1. The word frequency trend of the two is similar to the distribution trend of the annual publication volume, and both show exponential growth. It is worth noting that the keyword “Sustainable Development” exhibited a burst feature during 2006–2008 (bold line), that is, the keyword was widely used by scholars and appeared frequently in academic publications (Chen, 2006). This finding shows that the topic of sustainable development has been widely recognized by the Chinese Government, scholars, and society during this period, and began to form a rapidly growing research field. The burst features of sustainable development during 2006–2010 are closely related to China’s “Eleventh Five-Year Plan” policy to build a resource-saving and environment-friendly society during this period. Similarly, the keyword “Sustainability” has also become a hot spot for scholarly attention from 2009–2010 and showed a high word frequency (see Part A). After a brief retreat, the attention of this topic has also experienced exponential growth. In the past 15 years, China’s sustainable development research has received rapid development and widespread attention, which is reasonably validated in the above analysis.
———Insert Fig. 1 Here———
Furthermore, a descriptive statistical analysis is performed based on four indicators: country/region, institution, fund source and source journal, as shown in Table 3. Besides China (f = 11,970) itself, the country most concerned about China’s sustainable development is the United States (f = 1821). Australia, the United Kingdom, and Canada are also continuing to pay attention to China’s sustainable development issues. In addition to the mainstream role of English in the international academic community, the extensive network of academic exchanges and cooperation between China and the above four countries can also provide an explanation for this. In terms of institutions, five research institutions from China occupied the top-5 in the number of publications. The Chinese Academy of Sciences provided the most research evidence for sustainable development research in China (f = 2572), more than the sum of the contributions of the last four institutions. Many academic achievements in sustainable development research in China in recent years have benefited from diversified funding. Nearly half of the 12,653 publications used in this study were funded by the National Natural Science Foundation of China (NSFC), which are twice the number of funds that ranked 2–5. In terms of journals, journals such as Journal of Cleaner Production and Science of the Total Environment have received a wide range of submissions and have become the most important platforms for showcasing sustainable development research in China.
———Insert Table 3 Here———
3.2 Research hot-spot analysis
To a certain extent, keywords can reflect the core theme and knowledge base of an article (Chen, 2006). To this end, this paper conducts a co-occurrence network analysis of keywords. Based on 12,635 data items, parameters such as the node type of keywords, the cosine algorithm, and the top-50 threshold setting are used to conduct the co-occurrence network analysis of keywords. The time range we set is January 1991 to December 2021, and the time slice is 1 year. The visualization of the analytical results is shown in Fig. 2 which is divided into four parts, namely the keyword co-occurrence network (right side), Part A (burst strength > 10) and Part B (word frequency > 200), and the word frequency curve of the keyword “impact” (bottom right). The cooler colors in the picture are trending towards 1991, while the warmer colors are trending towards 2021. Based on 12,635 data items, we identified a total of 1262 keyword nodes and 6251 links among them, and the density of the entire network structure was 0.0079.
———Insert Fig. 2 Here———
As the keyword with the highest word frequency (F = 1782), the word frequency curve of “impact” is like the curve of annual publication volume, and both show a certain exponential trend. From the top-10 word frequency rankings (Table 4), the word frequency of the keyword “impact” is about three times that of the tenth keyword “urbanization” (F = 599). In addition, the word frequency of the keywords “management”, “model”, and “sustainability” all exceeded 1000 occurrences, all of which are the main topics of sustainable development research in China. In terms of the first appearance time of keywords, “sustainable development” was the earliest keyword in the top-10 word frequency rankings (1995), and relatively speaking, “policy” appeared the latest (2004). This reflects from the side that China has experienced ten years (1995–2004) from the introduction of the concept of sustainable development to the beginning of formation of relevant policies. As a comparison, we obtained the top-10 keywords with burst strength through burst detection. “trend” became the keyword with the highest burst strength (S = 32.13) and continued to play an important role during 2009–2017. The next “vegetation” has a strong representation during 2006–2015, and its burst strength (S = 31.96) also exceeded 30. Judging from the duration of burst, the keyword “winter wheat” continued to be the main topic in China’s sustainable development research during 2005 to 2015 (11 years). This, to a certain extent, explains the importance of agricultural development and grain output to the social stability of China (Gong, Yan, Wang, Hu, & Gong, 2009). Relatively speaking, the two topics of “greenhouse gas emission” and “driving force” have become relatively short-lived topics in China’s sustainable development research (2018–2019).
———Insert Table 4 Here———
The Timezone perspective is further used to interpret the co-occurrence network of keywords and the time zone distribution of key nodes (Fig. 3). We still use two indicators, Frequency and Burst Strength, to explore the distribution characteristics of keywords. Based on the threshold of word frequencies of more than 100 occurrences, we found that high-frequency keywords were concentrated between 1991–2010. Keywords such as “pollution”, “loess plateau”, and “land use” to “carbon emission”, “indicator”, and “technology”, to a certain extent, reflect that the focus of scholars is constantly shifting from a macro-perspective to indicators and paths (Y. S. Liu, 2018). There was a decline in research attention during 2011–2015. Subsequently, emerging topics and research perspectives such as “innovation” and “life cycle assessment (LCA)” began to appear from 2016–2021 (Umar, Ji, Kirikkaleli, & Xu, 2020). Judging from those eight keywords with a burst strength of over 20, their time zone distribution is like that of high-frequency words. Five high-strength burst keywords such as “trend”, “vegetation”, “basin”, “ecosystem”, and “winter wheat” were mainly concentrated during 2005–2010. Keywords such as “carbon emission”, “greenhouse gas emission”, and “innovation” also appeared among the emerging themes. From the above two perspectives, China’s sustainable development research has gradually shifted from macro-perspectives such as ecological pollution, vegetation protection, and geographical features to specific implementation methods and paths such as indicators, technologies, and innovation. In addition, China has paid great attention to climate change caused by greenhouse gases in recent years (Sarkodie & Strezov, 2018), which is of great significance to global sustainable development.
———Insert Fig. 3 Here———
3.3 Knowledge structure analysis
To elucidate the group characteristics of the keyword co-occurrence network and the knowledge structure of China’s sustainable development research, keyword-based clustering analysis (K-clusters) and Link Walkthrough are conducted on the network (Fig. 4). Figure 4 mainly includes two parts, namely the K-cluster view (center) and the Link Walkthrough diagrams (both sides). The Modularity value Q = 0.6079 (Q > 0.3), which is an evaluation index reflecting the modularity of the network, indicates the significant effect of the network clustering structure. The Silhouette value S of the evaluation index reflecting the homogeneity of the network is 0.8822 (S > 0.7), indicating that the clustering results have high reliability (Chen & Song, 2019).
Through K-cluster analysis, the keyword co-occurrence network of sustainable development research in China is identified as having 21 clusters. Except for three clusters (#18, #19, #23) which are far from the center, the network structure formed by the remaining 18 clusters is relatively concentrated, which all are closely related to the topic of sustainable development. In order to further identify the development context of China’s sustainable development research, we show the cluster evolution process from 1991 to 2021 through Time Slicing. It can be seen from the Link Walkthrough diagrams that the cluster structure diverged from the center to the periphery during 1998–2008, indicating that this stage is the key formation stage of China’s sustainable development research. During 2001–2005, the fields and topics of China’s sustainable development research have been greatly expanded, providing a foundation for the subsequent formation of the main structure. Therefore, we call this stage the “Expansion Area”. Interestingly, this phase coincides in time with China’s “Tenth Five-Year Plan” (2001–2005). The “Tenth Five-Year Plan” is an important milestone in China’s sustainable development research. This promotes the production of much scientific research and the publications, which in turn provides the basis for the formation of the Expansion Area.
———Insert Fig. 4 Here———
In the information table for the K-clusters, the cluster-IDs, size, annual average, and the top-3 cluster-labels based on the LLR algorithm are given (Table 5). As the largest cluster, “China” (#0, S = 141) closely fits the topic of this study, which verifies the effectiveness of data search from the side. Topics such as “sustainable development” and “energy consumption” involved in this cluster also outline the main directions of sustainable development research in China as a whole. The next three clusters, “water use efficiency” (#1), “soil organic matter” (#2), and “water resources” (#3), all have a size of over 70. Their topics all involve basic science and key resources related to sustainable development, such as soil, water, and plants. Relatively speaking, “institutional measures” (#18), “Delphi method” (#19), and “control strategy” (#23) become the three smallest clusters. It can be seen from the cluster labels that all three involve sustainable development strategies for policy control through measures such as indicators and prices. Combining the positions of the three clusters in Fig. 4, it can be judged that all three are independent issues in China’s sustainable development research.
All clusters are basically formed during 2001–2005 (except #7), which coincides with the Expansion Area in Fig. 4. Scholarly focus on topics such as dynamic perspectives and nutrient cycling makes “dynamics” (#7) the earliest cluster to form, and the only one to form before 2000. The subsequent two clusters of “loess plateau” (#6) and “township and village enterprises” (#10) make sustainable development research more in line with China’s scenarios and specific national conditions (Jia, Shao, Zhu, & Luo, 2017). The thematic differences between clusters represent the localization process of sustainable development in China to a certain extent. The three clusters “institutional measures” (#18), “China” (#0), and “industrial ecology” (#16), which were formed later, pay more attention to industries and policies (Mathews & Tan, 2011), as well as the shaping of sustainable development systems (B. Zhang, Bi, Fan, Yuan, & Ge, 2008). The analysis of cluster formation also proves that the establishment of China’s sustainable development research system is short-lived and rapid.
———Insert Table 5 Here———
3.4 Topic drift analysis
The process of topic drifts is identified based on a further combination of Timeline and Link Walkthrough (Fig. 5). Based on the Timeline of the annual slice, China’s sustainable development research has gone through three stages and two significant topic drifts:
(1) The Budding Stage (1991–1996). First, at this stage, scholars focus on the contradiction between rapid development and ecological carrying capacity in the process of urbanization, such as “pollution”, “land use”, “vegetation restoration” and “loess plateau” (Jia et al., 2017; Y. S. Liu, 2018). The research method is relatively simple, and scholars tend to adopt qualitative research methods such as conceptual analysis or case analysis. The Link Walkthrough indicates that the research scope at this stage is relatively focused, and there are no significant topic drifts, as shown in Part I of Fig. 5.
(2) The Primary Stage (1997–2015). With the approach of the Expansion Area, China’s sustainable development research began to experience a large-scale topic drift. This process is obvious in the Link Walkthrough, that is, many longitudinal connections extend from Part I to Part II. This drift lasted for nearly 20 years, mainly manifest as a shift from basic pollution issues in the budding stage to macro-themes such as climate change and biodiversity. The global ecological crisis caused by climate warming in the first decade of the 21st century has promoted human societal awareness of the need for ecological protection (IPCC, 2013, 2021), which has also brought about significant effects on China’s sustainable development. At this stage, scholars tend to use quantitative research methods (such as index, quantitative analysis, etc.) to estimate the specific impact of climate change on the sustainable development of China’s society and economy (Geng, Fu, Sarkis, & Xue, 2012). Different from the more focus on water resources protection in the budding stage, the research at this stage focuses more on the assessment of soil (X. B. Liu et al., 2010; Teng et al., 2014) and air pollution (Jim & Chen, 2008) and the analysis of potential impacts (such as research on cardiovascular, cerebrovascular, and respiratory diseases) (Hu, Cheng, & Tao, 2017). The resulting attention to environmental strategies at the macro- and meso-levels is also reflected in China’s environmental policy-making process (Lu et al., 2012). Based on the above analysis, we refer to the topic drift between the budding stage and the primary stage as the “Budding-to-Primary” transition.
(3) The Mature Stage (2016–2021). As the Chinese Government and scholars continue to pay attention to ecology, emerging hot-spots and controversial topics continue to emerge, and the process of topic drift follows. The drift process is obvious in the Link Walkthrough. That is, since 2016, many horizontal connections have been extended from Part I to Part III. Although the “smog problem” has become a social hot-spot in China as early as 2013 (Sueyoshi & Yuan, 2015), the in-depth control of air pollution has only become a topic of extensive attention and in-depth research by the Chinese government and academia at this stage (such as the “Blue Sky Defense” campaign). At this stage, scholars began to adopt a more systematic approach to scientific research and environmental assessment. For example, the wide application of data envelopment analysis (DEA) based on the multi-index system and the environmental assessment system based on the LCA provides effective tools for China to deeply analyze ecological carrying capacity and track carbon footprints (Z. L. Liu, Wang, Li, Bai, & Ma, 2022; Sueyoshi & Yuan, 2015; Sun, Wang, & Gong, 2022). In the above process, the wide application of environmental information disclosure system and emission testing facilities from the perspective of macro-environmental regulation has played a key role (Li, Huang, Ren, Chen, & Ning, 2018). In addition, the application of new technologies (5G, Internet of Things) and the practice of emerging environmental governance models (PPP, sharing economy, carbon trading market, green finance) provide a guarantee for China to move towards comprehensive and sustainable development (D’Orazio & Valente, 2019; Wang, Liu, Chan, Choi, & Yue, 2016; Xiaer, Tang, Yuan, Zuo, & Li, 2022; L. H. Zhang, Zhang, Duan, & Bryde, 2015). Based on the above analysis, we refer to the topic drift between the primary and mature stages as the “Primary-to-Mature” transition.
———Insert Fig. 5 Here———
3.5 Knowledge transfer analysis
Instead of topic drift, we try to explore the development and practice of China’s sustainable development research from a more macroscopic perspective. Based on the journal overlay-map analysis, the disciplinary drift paths of China’s sustainable development research from 1991 to 2021 are studied, which will help to identify the transfer process of knowledge units between disciplines, as shown in Fig. 6 which can be divided into three parts: Part A is the original layer, and the layers are marked with different colors representing the categories of the journals in the WOS database. Part B is the result of disciplinary drift after journal overlay-map analysis, and the line represents the transition from citing to citied literature in an article. Part C demonstrates the results of merging repeated paths through the Z-score method and then identifying the drift paths of key disciplines (Chen & Leydesdorff, 2014).
———Insert Fig. 6 Here———
Figure 6 shows that China’s sustainable development research has experienced a significant and large-scale disciplinary drift process from 1991 to 2021. Combined with the results of Z-score analysis, the disciplinary drift of China’s sustainable development research can be summarized into the following eight key paths: from the perspective of discipline sources, the most important source of knowledge in China’s sustainable development research is the “Ecology, Earth, and Marine” discipline. The discipline became the source of knowledge for the four key paths (4/8). Relatively speaking, the main destination of China’s sustainable development research results is “Economics, Economic, and Political”, which has become the knowledge attribution of the three key paths (3/8). Interestingly, the direct transition between the main source and destination is not the most important disciplinary drift path of China’s sustainable development research (the third one). There are still many indirect transfer or secondary drifts in the process of disciplinary drift in China’s sustainable development research. Based on the Z-score, the transformation from the discipline “Economics, Economic, and Political” to “Economics, Economic, and Political” has become the most important disciplinary drift path (Z = 5.099, F = 16,548). This is the drift path of the self-iteration of the “Economics, Economic, and Political” discipline, and at the same time makes it the discipline that is most closely integrated with China’s sustainable development research. Relatively speaking, the knowledge transfer process from the discipline “Ecology, Earth, and Marine” to “Environmental, Toxicology, and Nutrition” is the lowest scoring path among the eight key paths. Overall, seven of the eight key paths have undergone disciplinary transitions, which shows that China’s sustainable development research has experienced many disciplinary drifts during 1991–2021. Based on the above analysis, the key paths of disciplinary drift and the transfer process of knowledge units are further summarized (Table 6).
———Insert Table 6 Here———