­­­Discovering research trends of urban geology based on a bibliometric analysis

Keyword approaches and framework

The sequence of numerous UG definitions between 1950 till 2018 in various literature is presented in Table 1. Some UG definitions might be similar however they have different terminologies. This fact has led to a group formulation of preferred key concepts for UG that correlate with each other

From the definitions in Table 1, it was clear that the terms UG, environmental geology, and engineering geology are interchangeably used.^2,21,22 While the term urban was interrelated with the concept of “the city”.²³ The city collectively is defined as a concentration of buildings, roads, public and private spaces, people, conflicts, and common efforts that is administratively delimited.^21,22 An urban area has always performed with a wide range of city functions.²³ It is a settlement with a high population (where most of the population are not primarily engaged in agriculture, or where there is surplus employment), expanded beyond the administrative boundaries, and includes cities, towns even suburbs.^23,24

Moreover, the term city is frequently used to describe a metropolitan area, region, and urban agglomeration.²⁵ The Metropolitan area comprises of the urban space as a whole and its primary commuter,²⁶ typically formed around a city with a large concentration of people (i.e., a population of at least 1,000,000). On a larger scale, an urban agglomeration with 10 million or more is called a megacity.²⁷ These definitions suggest that the terms city (cities), urban, metropolitan (area), and megacity are interchangeable, depending on the context used.

The preferred key concepts (Table 1) also shows how the UG definition was often approached by the concept of urban planning and development.^4,25–27,30 Urban planning is the organized planning of the physical environment, where individuals live to create a healthy, reliable, and durable living space by providing safety in line with their social, cultural, and economic needs.³² Earth science factors (e.g., geology) are essential in planning for urban development initiatives. These factors address ground-related problems and other potential constraints on development.³³ The use of geology for urban planning and development has been applied to earthquake hazard vulnerability,³² landslide susceptibility and risk zoning,^33–35 seismicity,^36,37 geotechnical issues such as erosion and expansive soil,³⁸ and flood hazard.³⁹ To emphasise the importance of this matter in 2014 the International Association for Engineering Geology and the Environment (IAEG) XII Congress in Torino, published a series of books as part of its proceedings on environment, processes, issues and approaches, with volume 5 titled “UG, Sustainable Planning and Landscape Exploitation”.³¹ At present, the need for geology in planning and development in the urban areas are expected to increase due to the rapid population growth.

Multi-stage data processing

The journal articles related to UG from 1950 till 2018 were searched in the Scopus database on July 24, 2018. Scopus was selected because it has the largest single abstract and indexing database.⁴⁰ Additionally, Scopus is the leading citation source to journal articles, compared to other bibliometric data collection tools.⁴¹

The selection method comprises of three stages. In the first stage, keywords were combined with the use of Boolean operators such as “AND”, “OR”, and “NOT”, in the Scopus search engine. The selection of keywords was taken from the key concepts, that had appeared in various definitions of UG, as explained in the previous section. The first search was “urban geology”, which resulted in 167 documents. The use of quotation marks (“_”) was to search for the exact phrase as it appears in the articles. The second search was environmental AND geology*, which resulted in 14,087 documents. The * symbol was used to search for an alternate word ending, while AND was used to combine the searched phrases without it becoming an exact phrase. It meant that the search results may have been from documents containing the word ‘environmental’, ‘geology’ or both words. The third search was engineering AND geology*, which resulted in 25,303 documents. The fourth search was a combination of geology* AND urban OR city OR cities OR metro* OR megacity* AND planning OR development, that resulted in 4,798 documents. Some key concepts were truncated here as well with the use of the * symbol to obtain various search results (e.g., megacities, megacity for megacity*). OR was used to combine related terms or synonym for urban (i.e., city, cities, metropolitan, megacity). All four searches were thus stored in the search history. Finally, in this stage all four searches were combined as #1 OR #2 OR #3 AND #4. Sets of searches were combined using “OR” and “AND”. This combined stage resulted in 1478 documents. However, these results may include some irrelevant publications that had the searched keywords, that did not relate to UG.

The second stage involved the exclusion of document types, languages, and subject areas that were not directly related to UG. First, the search for the article type was limited, which resulted in 735 documents. Articles were then filtered for English language, reducing the resulted to 595 articles. The search was further narrowed down with the execution of subject areas such as “medical,” “physics,” “business,” “economy,” “arts,” “decision policy”, “chemical engineering”, “chemistry”, “material”, “mathematics”, “immunology”, “nursing”, “pharmacy”, “psychology”, “energy”, and “computer”. This stage produced 529 documents.

The third stage involved exclusion of topics that are too broad based on title, abstract, author keywords, and index keywords. Results from the previous step, including information on citation and abstract, author, and index keywords, were included when downloaded as PDF. Hence, to ensure the relevant content, 529 abstracts in the PDF format were scanned to determine further exclusion from the results. At the end, 285 articles were selected (See underlying data⁴²). The summary of the three stages, and their refined results are shown in Table 2.

Data analysis

In the analytical phase, 285 research articles were analyzed in terms of amount and time of publications, keywords, topics, and sub-topics. The authors used the Scopus feature, such as the metric article module, to statistically analyze the annual publication trend.

However, for the observation of the research trends, the authors used the clustering technique provided by the open licence software tool, VOSviewer version 1.6.16.⁴³ VOSviewer clustering was done based on the fractional-counting method on the keywords in relations to the clusters. Visualization was presented in each set by color (i.e., red, blue, or green), which indicated the group in which the cluster was mapped.⁴⁴ The clusters were further analyzed to answer the research questions.

Annual publication trend of urban geology-related research

The yearly distribution of the UG articles by publication is presented in Figure 1. The search timeline was set from 1950 (the year when the UG topic began to grow) until 2018, however, the years in which publications were found ranged from 1970 to 2018. Figure 1 shows a plateau in the number of publications between 1970 till 1981, with a slow increase in the number of publications from 1982 to 1997. A significant increase was observed during 1999-2016, as the number of research articles increased from 9 to 17. From the 285 articles, one was published in 1998, and 187 articles were published in the 2000s. This could be explained by the fact that global research (including research in UG’s field) declined due to the Asian financial crisis that happened in 1997-1998.⁴⁵ However, the UG concept emerged in the 2000s, especially after the 2004 Indian Ocean earthquake and tsunami disaster, to address urban resilience against natural disasters.

Figure 1. Annual urban geology research publication trend from 1970 to 2018 (as of the end of July).

Research topic interests in urban geology

Research questions such as, “What was the UG research topic from 1950 to 2018?” and “How did these research topic interests interact with each other?”, were answered with the use of the bibliographic data to construct a co-occurrence map in the VOSviewer software.⁴³ Several examples of similar analyses have been done in the field of general science and technology,⁴⁶ in geoparks,⁴⁷ in soil erosion,⁴⁸ and volcanic geomorphology.⁴⁹

For the data selection and thresholds, all keywords were divided into clusters with the minimum number of occurrences set at 15 keywords. Among the 2688 keywords, 42 met the threshold, which were presented as 42 nodes. Eck and Waltman suggested that in constructing the bibliographic coupling networks, “fractional counting” instead of the ordinary “full counting” methodology, can result in all publications to have the same counting portion.⁵⁰

The processed bibliographic data resulted in the keywords that were grouped into three clusters as presented by the VOSviewer in Figure 2. The three clusters were represented by three different colors, in which green represents cluster 1, red cluster 2, and blue cluster 3. The nodes in Figure 2 represent a term, and the node's distance reflects the relationship. Close distance between the nodes, reflect an intense relationship and a strong link between the two terms. Larger nodes represent a higher number of occurrences (high weighted). A summary of the clusters and terms are shown in Table 3.

Figure 2. VOSviewer keywords co-occurrence map. The visualization shows 42 terms that belongs to cluster 1 (green color), cluster 2 (red color), and cluster 3 (blue color).

Each term was connected to other terms by a link representing the relationship between the two terms. The stronger the link, the thicker the display line.⁵¹ All terms are quantified according to their occurrences and link strength, as shown in Table 4. The link strength indicates the strength of the relationship between the two terms and the total links between the nodes represents the sum of link strength of one node over others.⁵² As seen in Figure 2, there are several significant nodes on the map which indicate the most common terms. They are “Engineering geology”, “Geology”, “Urban planning”, and “Urban area”. These four terms were covered in cluster one and two.

The following sub-sectional outline in the three clusters represent the three research topics of interest, such as engineering geological hazard investigation and risk assessment in the urban area (EGR); social geology and urban sustainability (SGS); and urban hydrology and water management (HGW). In general, there are more research on EGR (42%), followed by SGS (33,7%) and HGW (24,3%) topics.

Cluster 1: Engineering geological hazard investigation and risk assessment in the urban area (EGR)

The green cluster (cluster 1) contains 14 nodes in which the keyword “Engineering geology” has the highest occurrence and total link strength. The node engineering geology showed thick lines connecting with most terms in all clusters, explaining the fact as to why UG research was mostly related to engineering geology. Other prominent terms in this area include “Geotechnical engineering”, “Subsidence”, “Eurasia”, and “Hazard assessment” (Figure 3).

Figure 3. Co-occurrence network of engineering geology as a keyword in VOSviewer.

There were 120 articles in this cluster. The articles were mostly related to hazard investigation and risk assessment on the underground civil planning (geotechnics), karst collapse and subsidence, landslide, seismic evidence for earthquake, and general geological hazard cases. All cases were viewed from the perspective of engineering geology. Almost half of the EGR research articles were focused on underground civil planning (geotechnics) cases in urban areas (56 articles). The most popular topic was tunnelling,^53–56 underground spaces,^57–59 and geotechnical modelling.^60–63 Case studies for these topics were mostly done in developed countries such as the USA (e.g., Los Angeles, New York, San Francisco, Boston), Japan (Tokyo), Canada (e.g., Metro Toronto, Ontario, Saskatchewan), United Kingdom (London), The Netherland, Singapore, etc.

The next most significant focus of the EGR articles studied were karst collapse and subsidence cases (22 articles), seismic evidence for earthquake cases (16 articles), landslide cases (14 articles), and other types of geohazards cases in general (12 articles). Research on karst collapse and subsidence were mostly done in European countries such as Italy, Spain, and Belgium.^63–65 While for landslide, the related topics are land-use and landslide,^66,67 landslide vulnerability and risk assessment.^68–70 Other issues related to seismic evidence for earthquakes are mostly on earthquake hazard cases, situated in Turkey’s urban area.^71–74 Eurasian Plate movement was the most frequently discussed topic in these earthquake hazard cases. In addition, some researchers analyzed the engineering-geological hazard investigation and risk assessment for all possible aspects in the urban area. We found that there were papers that discussed the role of engineering geology for building conservation.^75–77

The oldest publication listed in this cluster was an Indonesian study from 1970, which was on the application of engineering geology for the regional development and UG.⁷⁸ Previously, the main role of engineering geologists in Indonesia was to give advice to large civil engineering construction projects, in addition to increase the importance of human resources.⁷⁹ This might suggest that the initial idea of engineering geology as part of UG in Indonesia was not fully researched. However, there were a small number of articles in this cluster that were related to volcanic eruptions,⁸⁰ flood,⁸¹ and building stone decay and preservation⁸² (Figure 4).

Figure 4. Key topics in the EGR cluster and number of articles published.

Cluster 2: Social geology and urban sustainability (SGS)

Cluster 2 was represented by 14 red colored nodes in which the five highest weight and total link strength keywords were “Urban planning”, “Urban area”, “Land use”, “Urban development”, and “GIS” (Figure 5). However, the close nodes between “Environmental geology” and “Urban planning”, and the very thick line between “Geology” and “Urban planning”, can be used to explain how urban planning and geology were interrelated and what they shared, in the domain of environmental geology.

Figure 5. Co-occurrence network of keywords in cluster 2 (S.G.S.) using VOSviewer.

There were 96 articles in this cluster with the topic of SGS. The term social geology refers to the discipline of geology that studies the interaction between the geological environment and the social development, especially the influence of geological resources and risks on the territorial and social management of urban zones.⁸³ SGS included geo-environmental appraisal in the developing urban areas. It ranged from UG mapping for land-use planning (54 articles), GIS-based geo-environmental suitability assessment for urban land-use planning (19 articles), environmental monitoring, assessment, and landscape management (13 articles), monitoring, policy, and law for urban planning (10 articles).

The topic that received the most attention in this SGS cluster was related to UG mapping for land-use planning purposes.^84–88 The 54 articles on this topic were mainly published before the 2000s, with a slight decline after this period. This was before GIS (Geographic Information System) studies were well-developed and applied. Since the early 2000s, mapping for urban land-use planning has taken a GIS approach, instead of relying on field geological investigation. GIS-based geo-environmental suitability assessment for urban land-use planning has been the second major topic in this cluster.^89–92 Most of GIS approach in the study collaborated with AHP (Analytic Hierarchy Process) method.^93,94 Similarly, Ulfa, et al. presented the results of their study with the use of APH in geological research for urban land-use in Indonesia.⁹⁵ The third topic in this cluster was related to the environmental monitoring and assessment^96–98 and monitoring, policy, and law for urban planning^99–101 (Figure 6).

Figure 6. Key topics in the S.G.S. cluster and number of articles published.

Cluster 3: Urban hydrology and water management (HGW)

The third cluster was related to HGW. It was consisted of 14 nodes in which almost all terms contain “water” as part of the keyword (Figure 7). The most frequent relevant terms that appeared while linked, were “Storm sewers”, “Runoff”, “Stormwater”, and “Flood”. Stormwater was defined as rainwater that is runoff from land or built-up on surfaces such as roofs, driveways, pavements, footpaths, and road infrastructures, without entering the drainage system.¹⁰² The existing issues regarding stormwater are pollution and flood.^103,104 One of the best management practices to control stormwater pollution is developing a sewer system called storm sewer, expected to be different from wastewater sewer.¹⁰⁵ Moreover, storm sewers can be a solution for reducing floods by minimizing the discharge rate from the urban catchment areas.¹⁰⁶ However, this is more relevant to the urban water (hydrology) management, which is a domain of civil or environmental engineering instead of geology.

Figure 7. Co-occurrence network of water management as a keyword in cluster 3 (H.G.W.) using VOSviewer.

There were 69 articles in this cluster (Figure 8), of which 34 articles were focused on stormwater management (including flood assessment and modelling, urban stormwater, and storm sewer). The second focus in the cluster was on articles regarding wastewater treatment, including water quality and geochemistry (21 articles). The third focus was on 14 articles concerning groundwater, which was the only cluster that matched with geology as a scientific topic.^99-102 While the first and the second focus emphasized more on water engineering or applied technology aspects.

Figure 8. Key topics in the H.G.W. cluster and number of articles published.

Research gaps and future studies

In the past 40 years the increasing trend of UG research was in line with urbanization, even though there was a slow trend before the 2000s. Events such as the Indian Ocean earthquake and tsunami in 2004 that led to the killing of thousands of people who lived in urban areas, were triggers that increased the research in UG in the years to come. In 2015 the initiation of Sustainable Development Goal (SDGs) engaged geologists to have a role in helping and ensuring sustainable foundations for future global development.¹¹¹ Among the agreed geological aspects in SDGs that were in line with this study are engineering geology, geohazard, hydrogeology, and geo-heritage.¹¹¹ Since then, many UG articles were trying to relate their studies to the sustainable development concept.^76,112,113 Therefore, the research and application of UG has been postulated as a promising approach for sustainable development goals, especially for the 11^th goal (sustainable cities and communities), since 55% of the world’s population in 2018 was estimated to live in urban areas.¹¹⁴ It is expected that UG research will have increased popularity among researchers in the future.

The results of this study found that there were more research articles on EGR than on SGS and HGW topics. Most of the current research mainly focused on engineering geology related to hazard investigation and risk assessment for underground geotechnical construction. The main focus of underground geotechnical research was on case studies in developed countries, specifically in the Metropolitan areas, such as New York, San Francisco, Tokyo, Toronto, London, Singapore, etc. It is because the demand for underground infrastructure as solutions for traffic is growing in Metropolitan cities. Recently, those related to natural hazards such as subsidence, landslide, and earthquake in urban areas have also received attention, however they were viewed from an engineering geology perspective. In the future, it would be interesting to explore and examine the influence and challenges of UG in developing countries.

As shown in Table 4, the term “land use planning” has the lowest link strength within the whole three clusters. Among other terms which have lower link strength was the planning and sustainable development. These results indicate the most significant gap in UG studies was the interaction between geology and the land-use planning studies, which are under the umbrella of SGS. It was also indicated that sustainable cities and the communities have not yet considered geology for measuring a successful goal. Approach methods using GIS and AHP or even SMCE (Spatial Multi-Criteria Evaluation) can be explored in the future studies of geology for urban land-use planning and development.

Discussions on flood hazard and urban hydrogeology in this article were minimal. Urban water-related articles which are covered in UG topics were mostly discussed in terms of quantitative water management, storm sewer, and stormwater pollution, which does not fit in the scope of geology, even though it is covered in civil engineering. Articles within the scope, published prior to 2000s, were still lacking quantitative analyses. However, they are in need of necessary geological information on the water condition (either groundwater or surface water) presently, in order to answer practical hydrogeologic management and engineering questions. As expected hydrogeologic science is not well suited for quantitative prediction, however, it is best suited for providing theoretical and basic scientific solutions for complex practical problems.¹¹⁵

Underlying data

Figshare: Discovering research trends of urban geology based on a bibliometric analysis

DOI: https://doi.org/10.6084/m9.figshare.14864895.v7.⁴²

The project contains the following underlying data:

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).