Introductory Chapter: Soil Moisture – Keyword Analysis – A Bibliometric Approach

1. Introduction

In the fields of agriculture, ecology, and environmental science, soil moisture stands as a crucial factor of significant importance [1, 2, 3]. The water content in the soil, commonly referred to as soil moisture, plays a pivotal role in shaping ecosystems, influencing agricultural productivity, and affecting various environmental processes [4, 5]. In recent years due to advancements in research and methodologies, there have been improvements in how researchers analyze things, and this has happened across various domains. This has influenced various domains of studies, either thoroughly changing the perspective or having a significant impact to give a new direction to the research. One area experiencing notable scholarly attention is soil moisture. The field of soil moisture has received significant scholarly attention, with over 69,932 publications spanning from 1943 to 2024.

To gain a deeper understanding, various methods have been employed. Among these methods, bibliometric analysis stands out for its ability to conduct a meta-analysis of the entire domain. This method has recently widespread and became an essential method to get to know a subject. Bibliometric analysis offers analysis of authors, citations, collaboration between countries, and analysis of keywords. This chapter presents a thorough analysis of keywords extracted from the Scopus database using RStudio. The focus is on the topic of soil moisture, aiming to gain a comprehensive understanding of domain through bibliometric analysis of author keyword.

2. Methodology

The study has utilized the approach of bibliometric analysis to comprehend the overview of the field, to get a coarse understanding of the underlying relationships between soil moisture and other domains, and to observe the shifts or changes in pattern in the field throughout the years. The bibliometric analysis is a computer-assisted quantitative analysis that is based on the mathematical statistics and has been used in many studies to comprehend the overall topic, for example, soil water, soil nutrient research, etc. [6, 7, 8]. Though there have been studies/literatures focused on soil moisture [9, 10], studies that involve analysis using software are relatively less, and studies which have analyzed it on a large dataset in the timeframe 1897–2024 from Scopus database are close to none. This provides a unique perspective to this study. The bibliometric analysis especially focused on keyword analysis has been implemented using the Scopus database and RStudio in this study that tries to unveil the current standing of domain among other disciplines.

2.1 Data acquisition and preprocessing

The database chosen to extract the data is Scopus. The Scopus database is one of the widely known databases other than the Web of Science and incorporates various indexed journals. The files are easy to extract and in a CSV format. The keyword chosen was soil moisture which resulted in 133,470 documents throughout 1897–2024. The spectrum of result was narrowed by refining the search result with the keyword “soil moisture” and choosing publications like articles, review, book, book chapter, letter, note, data paper, and editorial and short surveys in the English language only. The advanced query is shown below. The number of documents considered for the study is 47,112. The data were extracted on December 4, 2023 [11].

2.2 Search strategy

The specific query which was used to get the required dataset from Scopus has been shown below:

QUERY: TITLE-ABS-KEY (soil AND moisture) AND PUBYEAR >1999 AND PUBYEAR <2025 AND (LIMIT-TO (EXACTKEYWORD, “Soil Moisture”)) AND (LIMIT-TO (DOCTYPE, “ar”) OR LIMIT-TO (DOCTYPE, “ch”) OR LIMIT-TO (DOCTYPE, “re”) OR LIMIT-TO (DOCTYPE, “le”) OR LIMIT-TO (DOCTYPE, “no”) OR LIMIT-TO (DOCTYPE, “dp”) OR LIMIT-TO (DOCTYPE, “bk”) OR LIMIT-TO (DOCTYPE, “sh”) OR LIMIT-TO (DOCTYPE, “ed”)) AND (LIMIT-TO (LANGUAGE, “English”). The data were then used as input to process the information and collect the results from RStudio (Biblioshiny) [12]. The issues with duplicity in names have been solved by using full names instead of short names for researchers before entering the information in the software.

3. Results and discussion

3.1 Keyword analysis and clustering

The keywords are the essence of any research, study, or publication [13]. Keyword-based clustering is a process of grouping different keywords based on criteria that how many times they have been used together or the underlying relationship between them [14]. In the study, author keywords are selected as they are provided by the authors themselves and contain the core of their research. The main overview of the whole analysis of the study is shown in Figure 1. The timespan covered is 2000–2024 with the number of documents being 47,112. The number of authors in the study is 95,126. The percentage of international co-authorship in the selected study is 32.19% representing global collaboration. The total number of keywords is 69,874 with average citations per document.

Figure 1.

Overview of data. Source: compiled by the author using [11, 12].

3.2 Most frequent keyword

The most frequent keywords in the study have been chosen based on occurrences. The word with the highest number of occurrences is soil moisture (8144, 35%). The other keywords are drought (1225, 5%), evapotranspiration (1204, 5%), climate change (1495, 6%), remote sensing (1190, 5%), soil temperature (827, 4%), and soil water content (772, 3%). The tree map in Figure 2 represents top 25 keywords, their occurrences, and the percentage they occupy among the 69,874 author’s keywords. The most relevant keywords signify the research areas, disciplines, and studies that are related to the main keyword “Soil Moisture”, and if we focus on chronological development, it will be easy to assess the shift that has taken place in the field throughout the years. These different categories or tags are often coupled together, for example, the studies based on climate change, drought, and evapotranspiration have been used with the context of soil moisture the most.

Figure 2.

Tree map of keywords. Source: compiled by the author using [11, 12].

3.3 Cluster formation of keywords

The network visualization graph is a graphic representation of keywords based on several criteria:

1. In a literature or publication, when a keyword is paired up with another keyword, it is represented as the link of that keyword. The link represents the collaboration between different topics or subsets of a topic or in some cases the collaboration between domains (keywords are the representation of domains).

2. The frequency of the keyword, how many times it has been used to date in the domain, is represented as the occurrence of that keyword. This occurrence has been shown in the form of the bubble in Figure 3. The size of the bubble/dot denotes the frequency of occurrences, and if the frequency is higher, the bubble will be bigger and vice versa.

3. The thickness of links between keywords represents how often those keywords are paired up together. It helps in establishing a relationship between different subjects/topics, for example, in Figure 3. The keywords soil temperature and remote sensing share a heavy bond with soil moisture, unlike unsaturated soil.

Figure 3.

Network visualization of keyword-based clustering for soil moisture. Source: Compiled by the author using [11, 12].

For a detailed analysis, Table 1 below denotes the individual keywords and the underlying relationship between them in the author-suggested theme. The clusters have been formed with the help of “Walk-trap algorithm” by RStudio, and significant features like betweenness and closeness are used to determine the cluster.

Cluster (author suggested theme)	Node	Betweenness	Closeness	Page Rank
1. Water management in agriculture	Evapotranspiration	11.28186938	0.018181818	0.046420166
	Irrigation	4.765539854	0.018518519	0.02576167
	Water use efficiency	1.140555793	0.014492754	0.018749217
	Water balance	0.693612222	0.014925373	0.015332114
	Runoff	0.892322149	0.015151515	0.015632777
	Groundwater	1.260318758	0.015384615	0.014172587
	Infiltration	2.729763181	0.014705882	0.012668829
	Transpiration	1.030508096	0.014492754	0.015284994
	Water stress	0.513080389	0.014705882	0.011685795
	Yield	1.207066714	0.015151515	0.013372874
	Evaporation	2.024736795	0.015873016	0.014714286
	Photosynthesis	0.190704111	0.012820513	0.011209509
	Maize	0.474992872	0.014285714	0.011621234
	Modeling	1.189455599	0.015151515	0.010792917
2. Soil health and nitrogen cycle	Soil	1.92943644	0.014925373	0.013505362
	Soil respiration	1.395414672	0.014705882	0.019131313
	Biochar	0.455467692	0.013888889	0.008368732
	Nitrogen	0.86574047	0.014084507	0.011096739
	Nitrous oxide	0.156787223	0.012820513	0.008095699
3. Remote sensing and data analysis	Remote sensing	9.813727337	0.018181818	0.044473273
	Data assimilation	0.422598905	0.013333333	0.01841772
	Smap	0.293401992	0.013513514	0.015912399
	Machine learning	0.451501843	0.013513514	0.009116864
	SMOS	0.597331939	0.013513514	0.015278341
	MODIS	0.469448742	0.013513514	0.012193916
	NDVI	0.159185813	0.013513514	0.013253686
	Land surface model	0.069386855	0.012820513	0.012434864
	Tibetan plateau	0.116959246	0.01369863	0.008369577
4. Soil water content and hydrology	Soil water content	1.950061971	0.01754386	0.017064967
	Soil water	3.345231234	0.016949153	0.015992928
	Unsaturated soil	0.006822498	0.011235955	0.004110938
5. Climate change and hydrological impact	Climate change	16.96983934	0.018867925	0.045583387
	Drought	8.721182797	0.019230769	0.038109047
	Precipitation	3.758765249	0.018518519	0.031435819
	Hydrology	1.17075593	0.015384615	0.018162584
	Rainfall	2.284859369	0.016666667	0.016717456
	Soil properties	0.464717552	0.014925373	0.008659178
	Vegetation	1.07441294	0.015151515	0.01526518
	Agriculture	0.572552747	0.014925373	0.012647543
	Soil organic carbon	0.27364249	0.01369863	0.007998912
	SWAT	0.09870105	0.013157895	0.009177264
	Grassland	0.327476449	0.014492754	0.009714476
	Salinity	0.205690218	0.01369863	0.008160043
6. Soil temperature and regional variation	Soil temperature	7.327812546	0.016129032	0.03763695
	Loess plateau	0.221382883	0.013513514	0.0089884
7. Temperature and organic matter	Temperature	9.222672489	0.016129032	0.02495931
	Soil organic matter	0.438833007	0.012345679	0.00739461
8. Soil moisture monitoring	Soil moisture	359.2892401	0.020408163	0.195671498
	Soil moisture content	0.272509433	0.014285714	0.007678182
9. General moisture-related topics	Moisture	0.411922585	0.014084507	0.011803876

Table 1.

Clusters, nodes, and themes in keyword-based clustering.

Source: compiled by the author using [11, 12].

3.3.1 Cluster 1: Water management in agriculture

This cluster is named as “Water Management in Agriculture” as it focuses on various aspects of water-related processes in agricultural practices. The keywords within this cluster, such as evapotranspiration, irrigation, and water stress, are interconnected, representing a comprehensive view of water management in agriculture. Keyword within this cluster forms a network that emphasizes mainly on water use efficiency and managing water stress that is essential for increased production and sustainable agriculture [15, 16]. The theme centers around optimizing water use efficiency, irrigation techniques, and mitigating water stress to enhance agricultural productivity sustainably. The closeness centrality index indicates how close these terms are to each other in the network, while the betweenness centrality suggests their significance in connecting different nodes. The PageRank highlights the importance of these terms within the overall network, emphasizing their influence in the domain. The betweenness index of the cluster is in the range of 11.28–1.189. The closeness is between 0.01818 and 0.01515. The PageRank is between 0.04642 and 0.0107. Keywords such as evapotranspiration with a value of 11.28 serve as crucial connectors between different aspects of water management in agriculture, highlighting their central role in the cluster’s network [17].

3.3.2 Cluster 2: Soil health and nitrogen cycle

The theme in cluster 2, namely, “Soil Health and Nitrogen Cycle” represents its focus on soil composition and the nitrogen cycle. This theme delves into maintaining soil health and fertility through practices like biochar application and nitrogen management [18, 19]. The keywords within this cluster, such as soil, biochar, and nitrogen, are interconnected, indicating a comprehensive view of soil health. The betweenness index of the cluster ranges from 0.1568 to 1.9294, indicating the significance of these keywords in connecting different nodes within the network. The closeness ranges from 0.01282 to 0.01493, suggesting the proximity of these terms to each other in the network. The PageRank ranges from 0.00837 to 0.01913. Keywords such as soil respiration with a PageRank value of 0.0191313 are more significant in cluster context to soil health and highlight the significance of managing nitrogen levels and soil composition for preserving soil health and fertility [20].

Comprehensive analysis of Cluster 2 reveals the interconnectedness of soil health and the nitrogen cycle, highlighting the significance of practices like application of biochar and managing nitrogen in the soil to maintain soil health and fertility [18, 21].

3.3.3 Cluster 3: Remote sensing and data analysis

The “Remote Sensing and Data Analysis” cluster brings together terms related to advanced technologies in agriculture. This theme emphasizes leveraging technology to gather and analyze agricultural data for informed decision-making [22, 23]. Keywords like remote sensing, data assimilation, and machine learning suggest a focus on integrating data for analysis [24, 25]. The betweenness index of the cluster ranges from 0.06939 to 9.8137. The closeness ranges from 0.01282 to 0.01818. The PageRank ranges from 0.00837 to 0.04447. The keyword remote sensing has the highest PageRank rank value suggesting its significance, and the keyword also has the highest closeness centrality and betweenness centrality suggesting that it is bridging different concepts in the cluster. Cluster emphasizes on the importance of collection and analyzing agricultural data using technology like machine learning and remote sensing [26, 27].

3.3.4 Cluster 4: Soil water content and hydrology

The “Soil Water Content and Hydrology” cluster encompasses terms related to water content in soil and hydrological processes. This theme highlights the importance of monitoring soil water content and understanding hydrological interactions to optimize irrigation and mitigate water-related stresses [28, 29, 30]. Keywords like soil water, unsaturated soil, and hydrology suggest a focus on understanding and managing water within soil. The betweenness index of the cluster ranges from 0.00682 to 3.3452. The closeness ranges from 0.01124 to 0.01754. The PageRank ranges from 0.00411 to 0.01706. Soil water has the highest closeness, that is, 3.3452, and soil water content has the highest closeness and PageRank, that is, 0.01754 and 0.017064, respectively.

Cluster emphasizes the importance of monitoring and managing soil water content for optimizing irrigation practices and mitigating water-related stresses.

3.3.5 Cluster 5: Climate change and hydrological impact

The “Climate Change and Hydrological Impact” cluster focuses on the intersection of climate change and hydrological processes [31]. This theme explores the impacts of climate variability, including droughts, changes in precipitation patterns, and alterations in soil properties, on hydrology and agricultural productivity. Keywords like climate change, drought, and precipitation indicate a comprehensive view of how climate affects hydrology citation. The betweenness index of the cluster ranges from 0.09870 to 16.9698. The closeness ranges from 0.01493 to 0.01923. The PageRank ranges from 0.00799 to 0.04558. The keyword climate change has the highest betweenness, closeness, and PageRank in this cluster.

3.3.6 Cluster 6: Soil temperature and regional variation

The “Soil Temperature and Regional Variation” cluster focuses on the temperature of soil and potential regional differences. This theme explores how soil temperature variations, influenced by factors like geographic location and land cover, affect crop growth and soil health. Keywords like soil temperature and Loess Plateau suggest a regional perspective on soil temperature. The betweenness index of the cluster ranges from 0.22138 to 7.3278. The closeness ranges from 0.01351 to 0.01613. The PageRank ranges from 0.00899 to 0.03764. The keyword soil temperature in the cluster is the most significant out of two of them as it has high betweenness, closeness, and PageRank.

3.3.7 Cluster 7: Temperature and organic matter

The cluster “Temperature and Organic Matter” focuses on the relationship between temperature and organic matter in the soil. This theme delves into how temperature fluctuations impact organic matter decomposition rates and nutrient cycling processes within agricultural soils [1, 32]. Keywords like temperature and soil organic matter indicate a potential correlation between these factors. The betweenness index of the cluster ranges from 0.43883 to 9.2227. The closeness ranges from 0.01235 to 0.01613. The PageRank ranges from 0.00739 to 0.02496. The keyword temperature is the most significant in this cluster due to its high value in the selected metrics provided by RStudio.

3.3.8 Cluster 8: Soil moisture monitoring

The “Soil Moisture Monitoring” revolves around terms related to monitoring soil moisture. This theme emphasizes the importance of real-time soil moisture data for optimizing irrigation scheduling and improving crop water use efficiency [33]. Utilizing advanced monitoring techniques, agricultural productivity can be increased while conserving water resources [34]. Keywords like soil moisture and soil moisture content suggest a focus on understanding and measuring soil moisture. The betweenness index of the cluster is considerably high, ranging from 0.27251 to 359.2892, indicating its significance in connecting different nodes. The closeness ranges from 0.01429 to 0.02041. The PageRank ranges from 0.00768 to 0.19567. The main keyword belonged to this cluster, that is, soil moisture itself with the betweenness value as 359.2892, closeness as 0.2040, and PageRank as 0.195671.

3.3.9 Cluster 9: General moisture-related topics

This cluster covers a variety of topics related to moisture, as suggested by the name “General Moisture-related Topics.” Nodes within this cluster include terms like moisture, which might cover different aspects of moisture in various domains. The betweenness index of the cluster is 0.41192. The closeness is 0.01408. The PageRank is 0.01180.

3.4 Timeline analysis of keywords

Figure 4 shows the frequency of usage of keywords words over the years. After the year 2000, evident increase in the keyword soil moisture can be seen, followed by other relevant keywords. The graphs also show that there has been an increase on some particular keywords after 2014 like climate change and drought representing that the domain inclined more toward preserving the environment in the last decade.

Figure 4.

Timeline analysis of keywords. Source: compiled by the author using [11, 12].

3.5 Countries’ production over time

The countries with the highest number of publications are represented in the Figure 5. This includes countries like Australia, Canada, China, France, Germany, India, Italy, Spain, United Kingdom, and USA. The publications in time period (2000–2024) in Figure 5 represent the development of the field/domain in the various nations. The highest number of publications has been done by USA with 13,868 (till December 15, 2022), closely followed by China (13630) (till December 15, 2022) which is further followed by Germany (3305), Australia(2909), Canada(2832), United Kingdom(2709), India (2432), France(2221), Spain(2088), and Italy(1953). The publications in China overtook the USA in the year 2019–2020 and are the top performer right now among countries.

Figure 5.

Countries’ production over time. Source: compiled by the author using [11, 12].

4. Conclusion

The bibliometric analysis conducted in this study gives us useful information about how soil moisture research is changing over time. Author keywords from year 2000 to 2023 are analyzed using RStudio. Occurrence of keywords such as “climate change,” “drought,” “evapotranspiration,” and “remote sensing,” each exceeding 1000 times and underscores their significance and prevalence within the field of soil moisture. Furthermore, clustering of similar and connected keywords gives 9 clusters that reveal significant themes and trends within the field. The timeline analysis provides a historical perspective on how soil moisture research has become more important over time, especially since the early 2000s. Also, the increase in keywords like “climate change” and “drought” since 2014 shows growing emphasis on environmental sustainability and resilience. Since 2019, a shift in publication trend is seen with China overtaking USA. The present work offers a comprehensive overview of research in soil moisture, identifying key themes and trends that help in a deeper understanding of soil moisture dynamics and its implications for agriculture, ecology, and environmental science.