Keywords

1 Introduction

While there has been abundant research on renewable energy in developed countries, less attention has been paid to energy system decarbonisation in less developed regions, though the latter are more vulnerable to the adverse effects of climate change, show rapid growth in domestic energy demand, and are rich in natural resources critical for energy transition (Apfel et al. 2021; Vakulchuk et al. 2020; Overland et al. 2019). One such region is Central Asia, which makes a particularly interesting case study owing to its strategic location and resources (Vakulchuk and Overland 2021). After the collapse of the Soviet Union in 1991, Central Asia came to play an important role in the global economy as a source of fossil fuels, primarily oil and gas. This is also reflected in the literature (Vakulchuk 2016; Vakulchuk et al. 2022). However, given the growing global demand for clean energy and materials critical for the energy transition, the countries of Central Asia may be of increasing interest to researchers of clean energy (Vakulchuk and Overland 2021; Eshchanov et al. 2019a, b, c, d).

The aim of this chapter is to provide a comprehensive overview of the profile and trajectory of energy research in Central Asia, with a particular focus on the social sciences. The existing review articles on Central Asia have tended to focus on specific topics in the scenario and sectoral analysis (Mehta et al. 2021; Kaiser and Pulsipher 2007; Karatayev et al. 2021). None has systematically reviewed the literature on energy transition or renewable energy.

This study presents a systematic review of the energy transition literature published between 1991 and 2022 based on data extracted from the Web of Science (WoS) Core Collection database. Using a combination of descriptive statistical analysis and bibliometric visualisation techniques, we address the following research questions: (1) What sub-topics are covered in research on energy issues in the countries of Central Asia? (2) Can a shift from research on fossil fuels to clean energy be identified? (3) What are the most influential articles and research organisations in the field of energy research in Central Asia?

In the next section, we present our strategy for selecting publications, the data extraction and screening processes and visualisation methods. This is followed by our results section, which addresses our research questions, illustrating the empirical results with data visualisations. In the final section, we present our conclusions.

2 Methodology

We based our identification and extraction of publications for review on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement (Page et al. 2021). Our search covered social science publications written in English and listed in the Web of Science (WoS) Social Science Research indexes (SSCI, CPCI-SSH, BKCI-SSH) for the period from 1991 to 2022. Relevant publications were first identified through the WoS Core Collection database by applying the following Boolean search string to the title, abstract and keyword fields:

((TS=((‘renewable energy’ OR ‘solar energy’ OR ‘solar power’ OR ‘wind energy’ OR ‘wind power’ OR ‘hydropower’ OR ‘hydroelectric energy’ OR ‘biofuel’ OR ‘geothermal energy’ OR ‘geothermal power’ OR ‘power plant’ OR ‘agrofuel’ OR ‘bioenergy’ OR ‘green energy’ OR ‘clean energy’ OR ‘energy efficiency’ OR ‘energy consumption’ OR ‘energy transition’ OR ‘nuclear energy’ OR ‘fossil fuel’ OR ‘coal’ OR ‘decarbonisation’ OR ‘low carbon’ OR ‘petrol*’ OR ‘gasoline’ OR ‘oil’ OR ‘fuel’ OR ‘electricity’ OR ‘natural gas’) AND (‘Central Asia’ OR ‘Kazakh*’ OR ‘Kyrgyz*’ OR ‘Tajik*’ OR ‘Turkmen*’ OR ‘Uzbek*’))) AND PY=(1991-2022)) AND LA=(English)

A total of 542 publications were initially retrieved. This was subsequently narrowed down to 305 publications following a screening process including manual relevance checks and the limiting of articles to those published in peer-reviewed journals. The extracted WoS data included the article title, author(s), year of publication, name/country of the research organisation, keywords, citation information and list of references.

The final sample of 305 articles was exported in tab-delimited format to the VOSviewer software. VOSviewer is an application for creating maps and visualisations based on network data (Van Eck and Waltman 2022). It can identify networks among scientific publications, journals, authors, research organisations, countries and/or keywords. Objects of interest (e.g. articles) can be linked, for instance, by co-authorship, citations and bibliographic links. VOSviewer provides three different visualisation modalities: network, overlay and density. In this study, we use overlay figures. An overlay figure represents items with labels and bubbles (Van Eck and Waltman 2022). The greater the weight of the object, the larger the label and the bubble. The colour of an item is determined by its grade, in our case, the grade is the year of publication. By default, the colours range from blue (earliest years) through green to yellow (latest years) (Van Eck and Waltman 2022). See Fig. 1 for an example of an overlay figure.

Fig. 1
A word network. The left side has a small node for Turkmenistan and a larger node for Kazakhstan for years before 2016. Some of the words are natural gas, oil, transition, and resources. The right side has keywords such as economic growth, energy efficiency, and C O 2 emissions, after 2016.

Main topics in research on energy-related issues in the Central Asian countries (Overlay visualisation of co-occurrence of keywords using VOSviewer)

Full size image

One of the main limitations of this study is that it is empirically limited to the WoS database. In future research, it would be possible to cover a wider range of databases, though we do not expect that the results would differ substantially. Another limitation is that our dataset includes only papers published in English, while Russian has remained an important language for academic articles on Central Asia and there is also a growing number of publications on the region in Chinese.

3 Results

3.1 Key Trends in Energy Research in Central Asia (Research Questions 1 and 2)

We analysed the co-occurrence of keywords to assess the attention given to different topics and how priorities have evolved. The visualisation in Fig. 1 is based on the number of articles in which the same keywords occur. The rationale for this is that the topics that these words represent should be closely related and can illustrate research trends and patterns (Maier et al. 2020) in the popularity of topics over time. Figure 1 uses an overlay visualisation to represent 28 keywords that appear at least 6 times in our 305 articles (out of a total of 1296 keywords used in our analysis). The size of the circles indicates the total number of occurrences of a keyword, while the colour indicates the period when most of the occurrences were noted. VOSviewer automatically assigns a range of years to the period during which most articles were published. The colour of keywords of articles published before 2012 or after 2020 are converted to the 2012 or 2020 colour value, respectively. The distance between the words also indicates their co-occurrence within articles.

The colour coding in Fig. 1 indicates that there has been a shift from fossil fuel-related research to research on the clean energy transition. Thus, keywords such as ‘renewable energy’ occurred in 32 articles, ‘CO2 emissions’ in 28 articles, ‘environmental Kuznets curve’ in 11. All appear as yellow-green circles, indicating that they have been used more frequently in recent years. By contrast, the keywords ‘natural gas’ (11) and ‘oil’ (23) were used more often around 2014–2016, as indicated by the blue/green colours. Analysing the proximity of keywords to one another, we find that the more recent research on clean energy relates more to ‘energy efficiency’ (14 times), ‘energy consumption’ (14 times) and ‘economic growth’ (12 times), as indicated by their size and proximity to each other. By contrast, the keyword ‘hydropower’ is located far from these other keywords and is coloured purple, indicating that the more recent clean energy research focuses less on hydropower than on other topics. The figure also indicates a lack of publications on ‘solar’, ‘wind energy’ and ‘bioenergy’, as these do not feature in the figure.

Figure 1 indicates that Kazakhstan is the most studied country in the region from an energy perspective. Analysing the distance between the keywords, we can see that the word ‘Kazakhstan’ is located closer to the clean energy keywords than the other countries. This indicates that Kazakhstan is also the country in which renewable energy topics have been covered most in recent years. By contrast, other Central Asian countries are located at a greater distance from the ‘recently occurring’ keywords related to clean energy. The distance between keywords indicates that the main topics researched in relation to ‘Kyrgyzstan’ (6) and ‘Tajikistan’ (8) are ‘hydropower’ (6) and ‘water’ (6), while the colour indicates that these were published in the earlier period. For ‘Turkmenistan’ (8), the most commonly occurring keywords are ‘natural gas’ and ‘oil’, while for ‘Uzbekistan’ (11) it is ‘oil’, with colour also indicating earlier publication dates.

To identify key research clusters and hot topics, we analysed bibliographic links between articles with at least 20 citations. This resulted in 6 thematic clusters involving 60 articles (Table 1). Two articles are bibliographically coupled when they link to a common third work in their bibliography. The identified clusters are (A) Energy research from a macroeconomic perspective; (B) Energy security in Central Asia; (C) Natural resources of Kazakhstan; (D) Renewable energy and energy efficiency; (E) Energy geopolitics; and (F) Energy policy (Table 1).

Table 1 Central Asian energy research clusters
Full size table

According to the cluster analysis, energy research in Central Asia is most actively researched from the perspective of macroeconomics (Cluster A) and energy security (Cluster B). Cluster A covers the impact of natural resources, energy consumption and carbon dioxide on economic growth and trade (twelve articles in total); the energy-financial development nexus (three articles); and the environment (four articles, including Ansari et al. [2020], which revisit the environmental Kuznets curve). Most of these studies use the statistical method of panel data analysis and focus on multiple countries, both within and outside Central Asia. In Cluster B (energy security), the authors analyse political perspectives on energy corridors, natural gas resources, gas production and pipelines (six articles), and electricity reforms and energy security (eight articles).

3.2 Influential Articles, Authors and Research Organisations (Research Question 3)

Following Maier et al. (2020), we identified the most influential articles from our selection based on the number of citations (Table 2). Among the most cited articles, the three main research topics were the macroeconomic impacts of energy use; energy geopolitics; and climate change mitigation. The most cited article (Sorg et al. 2012) focuses on the impact of climate change on glaciation, the impact of runoff on water availability, and the implications for irrigation, industry and hydropower.

Table 2 The 10 most cited articles
Full size table

Figure 2 shows a significant increase in the number of peer-reviewed journal articles and citations between 1991 and 2021. The annual number of citations grew rapidly, reaching 1598 in 2021. The first article in our selection was published in 1992, and until 2007 only a few articles were published annually. The number of peer-reviewed journal articles has been steadily increasing since 2007, accelerating after 2017 and peaking at 41 per year in 2020.

Fig. 2
A bar and line graph compares the number of peer-reviewed journal articles and the number of citations from 1992 to 2021. The number of peer-reviewed journal articles is the highest in 2020 at around 40. The plotline has an inclining trend with the highest citations in 2021 at 1600.

Publications and the number of citations by year over the 1991–2021 period

Full size image

Figure 3 illustrates the number of researchers from their respective countries who have produced energy-related research on Central Asia. The size of a sphere is determined by the number of published articles from that country. The colour of a sphere is determined by the period that most articles were published (blue/purple represents the earlier period, and green/yellow is more recent). Figure 3 depicts 33 countries whose researchers have produced at least three energy articles (of the total of 63 countries whose researchers produced our 305 articles). The top contributors are from the US (79 articles), China (57), the UK (37) and Kazakhstan (34). However, there has been a clear dominance in recent years of China, Japan, Kazakhstan and Russia, while researchers from the USA, UK, The Netherlands, Germany and Turkey were more active between 2012 and 2016 (Fig. 3). This reflects a decline in international scholarly interest in the topic of petroleum resources in Central Asia from 2010 onwards.

Fig. 3
A word network depicts the maximum number of research published on energy in Central Asia. The United States of America has the largest node for the year 2012. China follows in the year 2018. India, England, Netherlands, and Australia have the most research published in the years 2014 to 2016.

Countries whose researchers publish the most research on energy in Central Asia (Overlay visualisation of co-authorship link between countries)

Full size image

The data also showed collaboration between the 33 research organisations that published at least three articles included in our selection (out of the 487 organisations that were involved in the 305 articles). The colour of the circles is determined by the year the articles were published (the most recent being yellow), with proximity indicating tighter collaboration. The greater the number of articles published by an organisation with a collaborating institution, the larger the circle. Thus, greater collaboration in recent years had been seen between Kazakhstan’s Nazarbayev University (10 articles and 143 citations), the Asian Development Bank Institute (Japan) (3 articles and 31 citations), the CAREC Institute (China) (3 articles and 7 citations), and Al-Farabi Kazakh National University (6 articles and 31 citations).

4 Discussion and Conclusions

The main objective of our study was to provide a comprehensive overview of trends in energy research in Central Asia. We carried out a systematic literature review and examined 305 articles published between 1991 and 2022 in journals indexed in the Social Science Research indexes, namely SSCI, CPCI-SSH and BKCI-SSH, of the Web of Science Core Collection.

Our bibliometric analysis revealed a trend in research publication topics moving away from fossil fuels towards clean energy in Central Asia between 1991 and 2022. Despite recent growth in the number of articles published on renewable energy in these countries, there is a lack of articles on solar, wind and bioenergy. The top contributors to the literature in recent years are from China, Japan, Kazakhstan and Russia, while researchers from the USA, UK, The Netherlands, Germany and Turkey were more active in the 2012–2016 period. The strongest collaboration between research organisations is observed between Kazakhstan’s Nazarbayev University, the Asian Development Bank Institute, the CAREC Institute and Al-Farabi Kazakh National University.

This data provides us with a picture of the academic research networks currently operating in the field of energy transition in Central Asia. Such analysis can help those who are seeking research partners or a research base, those who might want to fund productive collaboration, and those wishing to better understand the dynamics of current research. This study contributes to understanding the clean energy transition in Central Asia; however, for deeper analysis, we recommend exploring more databases and including Russian-language publications.