SubCity: Planning for a sustainable subsurface in Stockholm

With an expected increase in urbanisation and low-carbon transition efforts, the planning of cities is becoming more challenging, and societies need to rethink how urban infrastructures will be constructed in the future. There is a growing recognition that the use of space below the city will need to be significantly enhanced. However, once transformed, underground space becomes a permanent feature, and major metropolitan areas worldwide are gradually acknowledging the subsurface as a valuable, non-renewable resource, emphasising the necessity for long-term, comprehensive, and sustainable planning of its utilisation. Sweden, including the Stockholm region, has favourable geological conditions for building underground facilities and a long tradition of subsurface engineering. Despite these advantages, Stockholm lacks a comprehensive, long-term underground plan or strategy. For years, major subsurface projects have been driven by short-term needs, potentially hindering the optimal use of space below the cityscape. The overall purpose of this paper is twofold. First, we explore the nascent area of scholarly work concerned with the case of Stockholm ’ s subsurface. We do so by evaluating the current status and potential of urban underground planning in Stockholm municipality. Second, we seek to advance existing planning knowledge and practices concerning Stockholm ’ s subsurface by identifying several distinct but inter-related gaps and challenges that impede the immediate integration of urban underground space into strategic decision-making for the future of underground planning in Stockholm. We suggest that further research is necessary in several key areas to facilitate the effectiveness and sustainability of long-term urban underground use and planning in Stockholm City and its metropolitan area.


Introduction
The share of the world population living in urban settlements is expected to increase to 68% in 2050 (UN, 2019), putting further pressure on the already stretched metropolitan areas worldwide.The rapid expansion of sprawling metropolitan suburbs to accommodate housing, commercial, and transportation needs has given rise to the loss of open (green) spaces, unequal access to public services, persistent road congestion, and traffic pollution (Brueckner, 2000).Low-carbon transition efforts and potential climate change impacts will likely create additional challenges for the already stressed city management.
Meeting the dual challenge of the rising rate of urbanisation and lowcarbon transition requires that societies significantly and even radically rethink how urban infrastructures will be constructed and how the city subsurface will be used in the future (Broere, 2016).There is a growing recognition that the use of space below the city landscape will need to be significantly enhanced (Cui and Lin, 2016).It can be observed that a rising number of major metropolitan areas around the worldincluding Helsinki, Hong Kong and Singaporehave embarked on developing long-term sustainable planning and management of the urban underground space (Reynolds, 2020).The most prominent Nordic example is the City of Helsinki, which has developed its Underground Master Plan with the main objective of promoting the more efficient use of the city subsurface in the future (Rönkä et al., 1998;Vähäaho, 2014;Vähäaho, 2016).
Sweden's largest agglomerations are part of the increasing urbanisation trend observed worldwide (OECD, 2020).It is expected that the number of people living in Stockholm County (Stockholms län) -Sweden's main metropolitan regionwill reach almost 3 million by 2030 (Region Stockholm, 2020), and the population of Stockholm municipality (Stockholms kommun) is projected to increase by 10% in 2031and 20% in 2040(Stockholms stad, 2022b)).The projected influx of new inhabitants will most likely put significant strain on the Stockholm municipality (and its metropolitan region more broadly) in terms of finding available space for housing, transportation, storage and other vital urban infrastructures and functions.Suburban boroughs, such as Rinkeby-Kista and Farsta, may experience a substantial 40% population growth (Fig. 1).In contrast, Södermalm, situated in the heart of Stockholm, is expected to retain its status as one of the city's most densely populated regions with limited potential for further expansion, except vertically.A projected minor decrease in population by 5% may indicate a shift towards offices, retail and services as the residential market changes (Stockholms stad, 2022b).This suggests that one of the central and busiest areas, traditionally vital for bridging the city's northern and southern regions, will persist in presenting challenges for accommodating the additional flows of people and goods from the growing suburban zones, maintaining the pressure on the increasingly limited local land for infrastructure, transit, and urban services.Moreover, Stockholm's lowcarbon transition efforts and potential climate change impacts will likely place additional stress on the city's stretched possibilities for management and planning.Inevitably, Stockholm will have to turn its attention to the prospect of enhancing the use of underground space and carefully coordinating the use of this finite resource to accommodate these challenges in the future.
Sweden, including the Stockholm region, has favourable geological conditions for building underground facilities and a long tradition of underground excavation and mining (BeFo, 2018;Tengeborg and Sturk, 2016).As Persson (1998) observes, Stockholm's bedrock is regarded as an excellent quality engineering material for underground constructions.Moreover, the capital of Sweden is recognised for its complex subsurface railroad and motorway infrastructures for rapid transit, such as the 110 km long Stockholm Metro (Stockholms tunnelbana) with 47 out of 100 stations located below the city ground or the Southern Link (Södra länken) and the Northern Link (Norra länken) with underground road tunnels stretching more than 8 km in total.
However, despite these advantageous circumstances, the Swedish approach to planning and developing urban underground space (UUS) has been described as being driven by the "first comefirst serve" principle (BeFo, 2020), and this trend has been particularly observed in the Stockholm metropolitan area.In contrast to comprehensive underground planning strategies adopted by some cities worldwide, such as the City of Helsinki (Vähäaho, 2014;Vähäaho, 2016), Stockholm currently lacks any comprehensive and long-term underground plan or strategy.As a result, the city's short-term needs obscure long-term perspectives and hinder the optimal use of the underground space today and in the future.For years, major subsurface engineering projects have been largely driven by demands concerned with, among others, combating traffic congestion between expanding suburbs in the northern and southern parts of the metropolitan area and increasingly compact central zones of the city (Svallhammar, 2008), gradually forming the so-called "Swiss cheese" of various existing and scheduled underground infrastructures below the city ground.Moreover, as BeFo (2020) observes, geological information has been scattered, and there is a lack of an institutional framework and regulatory policies that would compel the submission of such data to a centralised database.
Although these challenges have been addressed in a handful of expert reports focusing more broadly on underground urban planning (UUP) in Sweden (e.g., Öberg and Sjöholm, 2019;BeFo, 2020;SGU, 2017), to date, there has been considerably limited scientific attention devoted to the planning of Stockholm's subsurface, as researchers have primarily focused on analysing surface-level planning, especially regarding transportation and housing developments.
The overall purpose of our paper is twofold.First, we aim to address this research gap by exploring the uncharted area of scholarly work concerned with the case of Stockholm's subsurface.We do so by mapping and assessing the status and potential of urban subsurface planning in the Stockholm municipality.We provide a brief overview of the bedrock geology, highlight notable underground infrastructures, and take stock of current planning practices in Stockholm.Second, by recognising that the underground is a space that, once transformed, becomes a permanent feature, we seek to advance current planning knowledge and practices associated with Stockholm's subsurface.We identify several distinct but interrelated gaps and challenges preventing the immediate integration of UUS into strategic decision making for the future of UUP in Stockholm.
The paper is structured as follows.In Section 2, we provide a brief review of the relevant literature on urban underground use and planning.In Section 3, we outline the methods and materials employed in this study.In Section 4, we explore UUS in Stockholm municipality, with a focus on geological conditions, existing and planned underground infrastructures, and planning rules and practices.Section 5 addresses gaps and challenges preventing the immediate integration of UUS into strategic decision-making for UUP in Stockholm.In the final Section 6, we conclude the paper by suggesting potential ways forward.

Advances in urban underground sustainable use
There is a rapidly growing scientific recognition of the value of the city subsurface and its significance for sustainable urban transition in the future.This is reflected in the burgeoning scholarly work concerned with the planning, management, and utilisation of UUS.The research can be grouped into three interrelated themes relevant to the case of Stockholm's underground use.

Advantages and opportunities
There is literature exploring the advantages and opportunities of underground urban use and planning for cities today and in the future (e.g., Bartel and Janssen, 2016;Besner, 2016;Bobylev, 2009;Broch, 2016;Goel et al., 2012;Parriaux et al., 2004;Reynolds, 2020;Romanovich and Simankina, 2016;Tengeborg and Sturk, 2016;Volchko et al., 2020;von der Tann et al., 2020).For example, Parriaux et al., (2006) argue that UUS offers "a large untapped potential that, if properly managed and exploited, would contribute significantly to the sustainable development of cities".In this work, underground space development is perceived as an important strategy to meet challenges that may negatively impact cities on the ground, such as traffic congestion, lack of space or overheating and flooding.It is argued that strategic use of the subsurface may lead to better space efficiency by providing additional space for urban development without encroaching on land above ground (Li et al., 2016a;Xu and Chen, 2021).This may help, amongst other benefits, to accommodate growing populations, provide more housing options, and create new green, commercial or industrial spaces.Moreover, some researchers emphasise the future potential of developing the urban underground for improving city resilience concerning climate change impacts and low-carbon transition efforts (e.g., Admiraal and Cornaro, 2020;Bobylev, 2009;Broere, 2016;Cui and Lin, 2016;Huanqing et al., 2016;Makana et al., 2016).Beyond urban structural considerations, there is also a growing recognition of the role of access to other subsurface resources, such as geothermal energy or aquifers (Hunt et al., 2016).For example, Broch (2016) points out that the advantage of utilising UUS does not only reside in traffic-related purposes (e.g., rapid transit below ground) and provides examples of how rock caverns and similar subsurface spaces are used for different infrastructures and functions in Norway.Beyond the displacement of less desirable functions to the subsurface, UUS is also seen as multi-functional and contributes to several issues simultaneously, for example, improving surface transit flow while also preserving historic districts (Bobylev, 2016).

Challenges to underground urban planning
We also identify work that highlights the potential barriers and challenges to underground spatial planning for future city development (e.g., Admiraal and Cornaro, 2016b;Bartel and Janssen, 2016;Delmastro et al., 2016;Li et al., 2016b;Volchko et al., 2020;von der Tann et al., 2018;von der Tann et al., 2020;von der Tann et al., 2021).The growing attractiveness and usability of the subsurface can result in multi-dimensional conflicts around the exploitation of UUS for different purposes (Bartel and Janssen, 2016).UUS is here recognised as a valuable "non-renewable resource" (Bobylev, 2009) that requires careful consideration in urban planning practices and incorporation in any city masterplan.Bartel and Janssen (2016) argue that "strategic, comprehensive and holistic spatial planning is required to deal with the increasing range of demands for the utilisation of underground space and to better manage associated conflicts related to space, responsibilities and priorities".
Planning for underground spaces, however, poses a unique challenge for urban planners and designers as it requires making the invisible visible, according to Admiraal and Cornaro (2016b).This includes taking into account the interdependence of underground space with geology, as well as visualising what interventions are currently present, what developments could be possible, and how they would function underground.Underground spaces are increasingly seen as the last frontier of urban development, waiting to be utilised by those who stake their claim first.However, the "first comefirst serve" approach to using underground space has resulted in many cities grappling with self-inflicted chaos beneath their surfaces and considering how best to manage it (Admiraal and Cornaro, 2016b).Another complication arises from the fact that subsurface governance tends to be complex (von der Tann et al., 2020).In this context, the absence of clear purpose and leadership in anticipating demand for the city's subsurface and creating values for underground urban use poses significant challenges to the city's sustainable development (Huanqing et al., 2016;Maring and Blauw, 2018;Qiao et al., 2022).
The lack of effective governance of UUS can also be attributed to the challenge of dispersed, incomplete or inaccessible geological data (von der Tann et al., 2018;Volchko et al., 2020).Mielby et al. (2017) identify that for urban underground planning, it is critical that the strategic management levels, organisation and decision-making components are supplied with appropriate subsurface information.This is not just a question of information availability but also the right data at the right scale and time as part of a clear collaboration between planners and geoscientists (Mielby et al., 2017).Whilst advanced "vertical" models tying surface and subsurface are much talked about, Mielby et al. (2017) found no cases of such integrated models in their study at the time.
Moreover, a broader perspective on subsurface services that benefit surface habitation highlights potential conflicts around the overexploitation of multiple resources, but also interdependent and externalised impacts (Li et al., 2016a).For example, groundwater exploitation may lead to increased subsidence and, thus, deterioration of built structures.Hence, careful consideration of complex underground facilities with multiple functions is more likely to be able to address complex urban questions (Cui et al., 2021).Effective management of UUS thus requires prioritisation and strategic decision-making (Maring and Blauw, 2018), which in turn has the potential to ensure a resilient urban future (Besner, 2016).
Various approaches to managing such conflicts through urban subsurface planning have been proposed; for example, the Deep City Method aims to optimise UUS in urban development across disciplines for long-term growth (Li et al., 2013).The method focuses on the required operational steps to develop long-term visions and planning approaches, for example, through institutional capacity building.Maring and Blauw (2018) propose a method of Asset Management for the Subsurface (AMS) which has the potential to give "insight into possibilities, costs and risks, but also the value of the subsurface" (Maring and Blauw, 2018: 396).The focus here is on supporting decision-makers regarding complex decisions; however, this first requires increased awareness of the subsurface's interconnected potential (Maring and Blauw, 2018).To effectively exchange data between geoscientists and planners for sustainable urban development, Mielby et al. (2017) propose the adoption of Geo City Information Modelling.This, however, requires systematic approaches to data collection and delivery, something which they find only occurs on a project-by-project basis.Following the broader application of ecosystem services for the protection of biodiversity and ecosystems in support of human activity, the concept of geosystem services has been developed to recognise the role of abiotic services.Geosystem services are proposed to enable more effective and sustainable governance of the subsurface (Van Ree and Van Beukering, 2016).
In a European-focused study, Hooimeijer and Tummers (2017: 169) argue that adaptations to "law and regulation, policy and visions, knowledge exchange, and design and construction in urban development" within existing planning fields could readily accommodate the integration of the subsurface into urban planning practices.These changes are, however, interrelated.As Hooimeijer and Tummers (2017) observe in the Dutch context, "if legal regulations regarding subsurface aspects were in place, and their implementation was demanded in policies and visions, then professionals in charge would include the subsurface in the planning process".

Socio-political dimensions
Finally, we find that there is rapidly growing research interest in exploring socio-political dimensions and visions of urban underground development (e.g., Dobraszczyk, 2019;Garrett, 2020;Graham, 2010;Graham, 2016;Harris, 2015;Melo Zurita, 2020;Melo Zurita et al., 2018;Pérez and Melo Zurita, 2020;Reynolds, 2020;Scott, 2008).This reflects a broader trend towards understanding the role of urban infrastructure in shaping social, economic, and political relations in cities.This literature focuses on exploring how the city subsurface is conceptualised and imagined.Rather than being a political nothingness framed solely by geological properties and technological solutions, the urban underground subsurface is recognised and studied as a contested and unfixed socio-political landscape that triggers a plurality of future underground city visions (Melo Zurita, 2020;Melo Zurita et al., 2018).In this sense, urban underground development can have significant implications for, among others, urban governance, public participation, and social inclusion (Elden, 2013).For example, decisions about where to locate underground infrastructure can have an impact on the accessibility of services and the distribution of benefits and costs across different social groups (Cui et al., 2021;von der Tann et al., 2021).Similarly, the design and management of underground infrastructure can have implications for the safety and security of users (e.g., Ceccato et al., 2013), as well as the environmental impact of urban development (e.g., Cui et al., 2019;Zargarian et al., 2016).It is acknowledged that a deficiency in both vertical and horizontal knowledge transfer can have an adverse effect on interdisciplinary and holistic decision-making (e.g., Hooimeijer and Tummers, 2017;Norrman et al., 2016).However, there is still a lack of attention to how current UUS praxis and knowledge co-production are constrained by broader conceptualisations.
Prevailing visions of the underground describe a service layer in which to hide "unwanted urban equipment" (Lambertucci, 2019).In 2000, a report published by the International Tunnelling and Underground Space Association (ITA) emphasised the role of strategic visions in enhancing the use of UUS (Clarke, 2000).However, little attention has subsequently been given to the specifics of such visions in the literature concerned with underground development.Adjacent literature in human and economic geography spheres emphasises the role of imaginaries in underground territory and politics.For example, McNeill's (2019) study of Singapore's underground development discusses the role of the state's visions and territorial interest in vertical volumetric expansion.

Methods and materials
Due to the exploratory character of our study, we employed two quantitative methods, stakeholder workshops and desk research, to assist us in acquiring insights into the largely unexplored realm of scholarly work pertaining to Stockholm's underground use and planning.
Our study is built upon the foundation of two stakeholder workshops that were conducted in September and December 2020.These two workshops, titled "SubCity: Future Visions of Urban Subsurface," brought together a total of 20 stakeholders.This diverse group of actors included representatives from Stockholm City, the City of Helsinki, the Swedish Geological Survey (SGU), the Swedish Transport Administration (Trafikverket), the Rock Engineering Research Foundation (BeFo), as well as industry experts, architects, and academic scholars.The main goal of those events was to map and evaluate the existing status and potential of urban subsurface planning in the Stockholm municipality.Our primary emphasis was placed on examining current knowledge and practices and identifying key steps and actions necessary to address the growing demand for and transformation of underground urban spaces.The workshops revealed numerous gaps, challenges and uncertainties facing UUP in Stockholm, both currently and in the foreseeable future.These challenges included, among others, the absence of comprehensive underground planning on the municipal and regional levels, incapacity and uncertainty in predicting subsurface demand, the absence of established urban underground planning practices, and limited access to geospatial information.
Our workshop observations and findings were further enhanced through desk research, which involved gathering and analysing empirical data from various available sources.This material encompassed legal documents, planning and strategy documents on the municipal and regional levels, expert and industry reports, as well as other publications collected from relevant actors and institutions, such as Stockholm City, Region Stockholm, SGU, BeFo, the Swedish Transport Administration and Extended Metro Administration (FUT).Furthermore, we conducted a review of academic papers related to urban planning in Stockholm and non-academic publications that delve into the history of Stockholm's development.
The data used for the production of the maps of Stockholm municipality's bedrock types and deformation, soil type and depth, as well as its topographic maps, were sourced from SGU, while the other layer of information on the maps about administrative boundaries, roads/rails names and locations were sourced from the Swedish Land Survey (Lantmäteriet).These data sets were overlaid on base maps obtained from Esri archives, and all population statistics were calculated based on the numbers sourced from Stockholm's City publicly available database.All these data were received in raw format and reprocessed to fit the paper's context.
Several limitations were encountered during the course of this study in collecting and examining empirical materials.The subsurface aspects were conspicuously absent in Stockholm's planning and strategic documents on both local and regional levels, whereas industry reports and expert literature were usually narrowed down to specific construction projects.We also observed a considerable lack of attention to Stockholm's subsurface in academic literature, and, to a large extent, we had to rely on non-academic expert publications.Moreover, the majority of relevant material was available only in Swedish, thus limiting the audience outreach.Finally, we have also encountered limited access to geological data in creating the paper figures.

Urban underground planning in Sweden: The case of Stockholm's subsurface
In this section, we provide a brief overview of the bedrock geology of Stockholm City and its deformation characteristics to highlight the geological constraints for UUS.We also highlight notable underground infrastructures and take stock of current planning practices in Stockholm.

Bedrock geology of Stockholm municipality
Much of Sweden is underlain by a hard rock basement of Precambrian Shield rocks formed approximately 2800-545 Ma.In places, younger metasedimentary rocks (545-55 Ma) overlay these.The bedrock in the broader Stockholm area was formed during the Svecokarelian orogen (2000-1800 Ma), similar to the formation period of most of the bedrock in eastern Sweden.Within this formation period, different lithotectonic units exist based on differences in bedrock origin, structural, metamorphic, and chronological development and are usually bounded by regional deformation zones (Wahlgren et al., 2019).This property of heterogeneity may add a layer of complexities during regional underground infrastructural development.
The bedrock in Stockholm municipality belongs to the Bergslagen lithotectonic unit and comprises mainly metasedimentary rocks (greywacke in Fig. 2) composed of veins of light quartz, feldspar and darker mica minerals.These are supracrustal rocks which metamorphosed after deposition and solidification of sediments on existing crustal bedrocks around 1910 Ma ago.Also predominant are metagranitoids of granitic to granodioritic as well as granodioritic to tonalitic compositions (shades of pink in Fig. 2), mainly in the western to central regions of Stockholm municipality.These are fewer in occurrence and can be observed as elongated bodies around Kista, Spånga, Hässelby, Farsta and Årsta areas within Stockholm municipality.However, the granitic to granodioritic compositions are more prevalent and form wider bodies along Hässelby through Bromma and Liljeholmen to Södermalm areas.Small bodies of mafic rock types varying from metabasites to gabbro-diorite (dark pink and yellow in Fig. 2) are scattered within the central metagranitoids.
Also prominent in Fig. 2 are amphibolite (diabase or dolerite) dykes, which are more or less oriented parallel to the dominant fracture and fault zones (black and red dashed lines in Fig. 2), especially in the east and south.These dykes may usually form poor-quality zones in terms of geotechnical properties within the bedrock.

Bedrock deformation of Stockholm municipality
An overall analysis of deformation in the rock mass between welldefined zones does not exist for the Stockholm area (Wahlgren et al., 2019).However, local deformation information from different projects in and around Sweden is available at SGU. Deformations can be classified into ductile, brittle-ductile and brittle types based on any combination of deformation depth, deformation temperature, prevalent stress regime, etc.The brittle deformation types can be observed in Fig. 3.The brittle fractures are predominantly sub-horizontal, extending northeast to north-northeast, while the unclassified deformations have more varied orientations.For instance, four fracture sets (northwest, northeast, north-south and east-west) with steep to horizontal dips have been reported during the Citybanan projects along the Södermalm area (Armengol, 2012).Meanwhile, fracture orientations from drill cores from the same area showed east-west with steep and gentle dips, westnorthwest with steep dips and east-northeast with moderate dips fracture sets.This highlights how easily geological conditions can vary within a project zone and thus require continuous, detailed characterisation and avoidance of inaccurate knowledge transfer.
Some of these brittle fractures are long and continuous, showing regional trends, mainly from the northwestern to the southern parts of Stockholm municipality.That these represent zones of weaker rock is evident as they often delineate borders of district areas (stadsdelsområde) and mountainous regions (Fig. 3).For example, in the eastern part, two continuous brittle fractures delineate the border of Norrmalm.Another set, albeit discontinuous, runs along the borders of Södermalm through the border of Långoholmen island.The two faults mapped in this area (red dash lines in Fig. 3) are both in the southern part and have subhorizontal strikes.One of these appears to be a part of the long, continuous brittle fracture running through the Hägersten-Älvsjö district area.This suggests that these brittle regional fractures may very well be a part of the classified faults system.Also, notable in Fig. 3 is the exposed fold formation delineated by two fractures, with one at the limb running through Hägersten-Älvsjö and the other at the limb running through Långholmen-Södermalm border.
The thickest quaternary soils deposits (6-15 m) within the Stockholm municipality are also commonly found in these weakness zones.This has a direct consequence for the average depth to bedrock in Stockholm municipality.Following the retreat of the ice sheet during the last glaciation, an esker (green in Fig. 3) is observed as a long and narrow glacial sediment ridge running through Norrmalm and Södermalm in a north-westerly direction.Wahlgren et al. (2019) report that some of these eskers, common in south-central Sweden, can be as thick as 50 m and have clays (also visible in Fig. 3) covering their sides.
Although there are inherent structural complexities within the bedrock of Stockholm municipality that would have a direct consequence for underground construction, careful consideration of these deformations (e.g., the presence of poor-quality dykes or soil at depth) in designing underground construction projects may minimise their consequences.This is supported by the numerous successful underground infrastructures in and around Stockholm.

Examples of existing and planned underground infrastructures in Stockholm
Situated on the bedrock regarded as an excellent quality engineering material (Persson, 1998), the capital of Sweden has, during the last century, made use of its favourable geological conditions for underground constructions and facilities (Fig. 4).However, the utilisation of Stockholm's subsurface has been predominantly driven by strategies aimed at resolving various pressing needs and challenges, such as traffic congestion and even military security.The Swedish "first comefirst serve" approach to planning and developing urban underground is particularly evident in the Stockholm metropolitan area, which lacks any comprehensive underground plan or strategy.Consequently, different existing and scheduled underground infrastructures form the so-called "Swiss cheese" below the city surface.Here, we briefly highlight key developments in Stockholm's underground use.
Sweden's capital is known for its 110 km long Stockholm Metro (Stockholms tunnelbana), also known as "the world's longest art gallery" (Cox, 2015).The decision to build the Metro was driven by several factors, including the city's expanding population, the need for a reliable and efficient public transportation system, and the desire to reduce traffic congestion, particularly between the northern and southern parts of the city (Svallhammar, 2008).Over the years, this rapid transit system has grown to include over 100 stations, of which 47 are located below the city ground, making it one of Europe's most extensive subway networks.Three lines -Green, Red and Blue (Gröna linjen, Röda linjen, Blå linjen) -are currently in operation (Fig. 4).
The Metro lines provide a fast and efficient transportation link between the city centre and the continuously growing suburbs, particularly in the south and northwest of Stockholm.The youngest of the three, the Blue Line is the deepest metro line in the city and one of the deepest in the world, with 19 out of 20 stations located below ground.Due to its average station depth, the line required significant engineering and construction efforts, including the use of tunnel-boring machines and other advanced techniques (Svallhammar, 2008).Moreover, the Stockholm Metro is actively expanding to fulfil the city's increasing transit demands.The Blue Line's proposed extension to Nacka's southern suburb includes Sofia station, situated 100 m underground, marking it as the Metro's deepest stop (SLL, 2017).There are also plans to build a further line from Arenastaden to Odenplan, which will connect with today's Green Line to the south, as well as an additional new line with six stations between Fridhemsplan and Älvsjö (SLL, 2016;Region Stockholm, 2016).The Stockholm Metro is supplemented by a six-kilometrelong commuter rail tunnel, the Stockholm City Line (Citybanan), completed in 2017 (Trafikverket, 2017).The tunnel was built to relieve congestion on the rail network and improve connectivity between the southern suburbs and central Stockholm (Fig. 4).It has also provided new opportunities for transit-oriented development in the areas surrounding the new station.The City Line includes two new underground railway stations, of which Stockholm City Station is situated between 35 and 40 m below ground (Trafikverket, 2017).
Another example of a rapid transit system utilising UUS is the large Stockholm Ring Road (Stockholms ringled) project suggested as early as the 1960's to address concerns regarding traffic congestion and air pollution (Abrahamsson, 2004).The new road project would provide a faster and more direct route for motorists to travel around the city, reducing traffic congestion in the centre and improving the overall flow of traffic (Fig. 4).A more concrete plan for the complex of motorways and tunnels around Stockholm was proposed in the 1990's as part of the Dennis Agreement to accommodate Stockholm's future increase in traffic demand (Sturk et al., 1996).The planned Ring Road would consist of southern, northern and eastern links, with the western section (Essingeleden) already in place since the 1970's.Completed in 2004, the Southern Link (Södra länken) is Sweden's longest (4,6 km) and most complex road tunnel system.The Northern Link (Norra länken) had to be constructed in stages, with the Karlberg-Norrtull section opened in 1991.However, in 1997, the project plan was cancelled due to the construction's negative impact on the Royal National City Park.After several revisions, the work was given the go-ahead in 2006, and the second section of the link, Norrtull-Frescati-Värtan, was opened in 2014.The complexity of the Northern Link project can be attributed to the fact that it intersects with multiple existing metro lines at various points (Rosengren, 1993;Trafikverket, 2016).
In addition to underground transit systems, Stockholm showcases other instances of UUS, like underground air raid shelters built during the Cold War to safeguard against Soviet nuclear threats (Fig. 4).Constructed to endure nuclear blasts, these shelters offered protection from radiation and stocked essential supplies.An example is the expansive Katarinaberget shelter beneath Södermalm, with a vast 15,900 square metre area, accommodating 20,000 individuals with air filtration, generators, and medical facilities (Roth, 2013;Westerberg, 2022;Okla, 2022).However, it closed in 2016 for modifications tied to adjacent construction, prompting debates over its decommissioning amidst recent geopolitical instability (Westerberg, 2022).Klara (Klara skyddsrum), another bunker built in the 1960's, lies beneath central Stockholm, once capable of housing 8,000 people.Presently, many Cold War-era shelters are either abandoned or repurposed as parking due to deterioration (Okla, 2022).Another illustration of repurposing Stockholm's Cold War underground structures for modern use is the decommissioned command post, Pionen, under Södermalm (Fig. 4).Initially designed for civil defence in war, it is now a data centre operated by Bahnhof since 2008.The data centre, situated 30 m underground in a stable two-billion-year-old granite region, offers optimal conditions (Veel, 2017).Similarly, Södersjukhuset (SöS), a major emergency hospital located in Södermalm, repurposed an underground cavern originally created during its wartime construction.Renovated in 1994, it now serves as a disaster and emergency medical centre, capable of accommodating up to 160 patients during war and acting as a shelter for 3,200 individuals (Ask et al., 2018;Glömd historia, 2016).
Finally, the Royal Library (Kungliga biblioteket) in Stockholm's Humlegården Park exemplifies how the surrounding green area and the building's historic architecture were preserved by turning to the underground for additional storage capacity (Lindblom et al., 1995).In the 1990's, the library's shelving space was substantially extended by connecting the on-ground facilities with two underground rock caverns of around 9000 square metres in size, each having five floors (Stugart, 1994).Today, the underground magazine can be visited via the library's online tool, available here: https://360magasin.kb.se/.
Whilst Stockholm's underground space utilisation is diverse and relatively extensive, few facilities are or were planned in coordination with each other.Different administrative bodies are responsible for each sector's development and can also differ depending on whether a project is planned at the regional or municipal scale and whether a project concerns an existing facility or a new one.

Planning aspects
Sweden is seen as one of the volumetric world leaders in underground development (Bobylev, 2016;BeFo, 2020).Thus, it is somewhat surprising that despite efforts towards greater integration, there is currently "limited opportunity" (SGU, 2017) to plan for underground facilities or to do so in a long-term perspective.In this section, we identify key challenges in planning praxis currently preventing comprehensive management of the subsurface in Sweden and, more specifically, Stockholm (Table 1).First, underground space use mainly occurs without formal strategic coordination (BeFo, 2020: i) and can be quite disconnected from surface land use planning.This is despite a wide variety of potential uses outlined above, which also have significant incompatibilities (BeFo, 2020) if not planned carefully.Second, Sweden's subsurface is governed by inappropriate property rights and inconsistent application thereof.This limits the potential instigation of a masterplan at the moment (SGU, 2017).Third, the perspective of the underground as a "precious, multifunctional and finite resource" (BeFo, 2020: v) is much discussed for its governance potential, but it lacks a clear means of application in a subsurface governed by sectoral legislation.

Land use planning in Sweden
Land use planning in Sweden is governed by municipalities enabled through the national regulations of the Planning and Building Act (PBL) ( Öberg and Sjöholm, 2019).However, it is worth noting that the regional administrative level takes on various responsibilities, including public transport planning, which can have implications for using the

Access to data/drawings
Publicly accessible: maps, property demarcations, overview plans, detailed plans, and more Access to comprehensive planning information for rock chambers, tunnels and other facilities is often restricted by the Protection Act (2010:305).This limits access to details or complete picture of infrastructure, even within government organisations.
urban (sub)surface in constructing transport-related infrastructure (Paulsson and Isaksson, 2019).A key consequence of the PBL is that municipalities must produce an overview land use plan (a masterplan).This covers the entire municipality, providing strategic-level guidance on how land and water should be used.Such plans are legally required, but their contents are not legally binding ( Öberg and Sjöholm, 2019).More detailed overview plans concerning a specific geographical area, along with feasibility studies, for example, can also be important steering or guidance documents with regard to land use planning.The next level of planning information is normally the so-called "detailed plan".The detailed plan is legally binding and defines the extent of new constructions, alterations to existing properties, land reservations for specific uses, etc.Additionally, there are area plans which can also be binding and can be used to capture land not covered by the overview or detail plans, sitting between the two major plan scales.Stockholm's overview plan does not mention the subsurface at this time.
Facilities under the surface are, in principle, covered by the PBL, although very limited guidance is provided on underground use beyond that underground facilities shall not impede the use of the surface (Planoch bygglag 2010:900 2 Kap. 8 §).Section 4 of the PBL notes that a detailed plan can be used by municipalities to define the extent of a building above and underground.However, this does not appear to help with strategic planning as the level of information is often too detailed and occurs in relation to specific projects rather than at the strategic decision stage.The use of detailed plans for UUS is complicated by the lack of specific guidance on how detailed plans should be applied to UUS, despite the principle that detailed plans should be updated to reflect tunnels and rock chambers (Adolfsson and Boberg, 2015).
Plans (especially overview plans, which include strategic prioritisation) should, in principle, be referred to in order to resolve space allocation conflicts that may arise, yet currently, plans do not cover the full subsurface subject matter ( Öberg and Sjöholm, 2019).Hence, there is a clear need for subsurface planning covering both underground and surface implications (BeFo, 2020), but there is limited agreement in the literature on how current legislation should be amended or supplemented to enable this.
Most underground functions, especially larger infrastructures, are planned through sectoral workflows.Transport bodies, such as the statecontrolled Swedish Transport Administration (Trafikverket) and regional Extended Metro Administration (Förvaltning för utbyggd tunnelbana), hold particularly prominent positions and make strategic assessments around transit capacities rather than the municipalities, albeit in consultation and in reference to municipal growth plans (see examples in Öberg and Sjöholm, 2019;Linnerborg, 2020).
Planning permits are required according to the PBL for certain construction types at the project scale, with key sectoral exceptions.This means planning permits are needed for telecommunication, wastewater, and district heating tunnels, for example.Tunnels and rock caverns which are intended for roads, railways, metro systems, or mining are not governed by the PBL.However, facilities tied to these functions, such as station buildings and commercial areas, require permits.An underground construction project may require other types of permits in certain situations, e.g., where groundwater will be affected or projects with significant environmental impact, such as a large road or a railway project ( Öberg and Sjöholm, 2019).These are touched on in the next subsection.

Ownership and property rights
As in many other countries, planning in Sweden is based on property formations.Sweden's surface is formally divided into properties, which are traditionally defined in the horizontal plane on the ground, with limits extending infinitely upwards and downwards ( Öberg and Sjöholm, 2019;Boverket, 2004)."Cuius est solum, eius est usque ad coelum et ad inferos" (Latin for "the owner of the land, owns it up to heaven and down to hell") applies (BeFo, 2020).
That said, since 2004, it has been possible to define 3D properties with a vertical delimitation.This is enshrined in the Land Code legislation (Boverket, 2004).Application may not be so straightforward, however, and in light of the new 3D rules, there is a need to better define and more consistently apply these concepts and terminology (Boverket, 2004).The risk of confusion arises as there is a provision in the Land Code and in the Property Development Act, which means that the laws within these documents apply both to land and to spaces not directly connected to surface land.The Planning and Building Act, on the other hand, has no provision for this equivalence.As such, an air volume (or it follows, a subsurface volume) cannot be subject to requirements of the PBL and thus is not readily subject to planning of its development.
Agreements and easements are the main means of currently enabling underground development regarding ownership: Responsibility falls on the surveying authority (lantmäterimyndigheten) to assess the most appropriate property formation methods (Boverket, 2004).Preference is given in the first instance to the traditional property, which should be coerced into an appropriate formation with the use of usufructs and easements.A usufruct is a partial solution that does not offer the facility owner certainty as they are not unconditional rights and are timelimited (Victorin, 2004).Easements are more certain, legally speaking.However, they are tied to specific properties and also to specific uses (Victorin, 2004), which does not readily facilitate future adaptation of the subsurface space, such as subdivision of space (Paulsson, 2007).
3D formation can only be used once it has been demonstrated that traditional property instruments cannot meet the purpose.As such, easements must first be discarded, which may be difficult as they are currently seen as capable (if perhaps not ideal) of managing underground developments.An example offered of a permitted case for 3D rights is that of a rock chamber that is disconnected from other properties, which is intended to be used independently (Boverket, 2004).A further issue around 3D property formation is that property development in accordance with the detailed plan is restricted to a specific use, thus meaning that the detailed plans must take account of future underground functions if they are, in turn, to be permissible (SGU, 2017).
The role of property expropriation by the state should also be mentioned, for it is a technique also common in other countries (e.g., compulsory purchase in the UK) that enables large infrastructure developments crossing multiple properties.Under the provisions of the Real Property Formation Act, the Joint Facilities Act and the Utility Easements Act, the municipality can expropriate land for public facilities such as a road tunnel.
Given the projected increasing development beneath the surface and the benefits of a more comprehensive and interconnected UUS, these legislative barriers appear to significantly limit successful masterplanning of the subsurface.We must ask how 3D space can be planned and managed adequately if the means of ownership and regulation are incompatible.

Legislation by sector
There is a lack of overall governance and guidance regarding the underground and fragmented policies (SGU, 2017), often instigated at the municipal level rather than the national level.The underground is invisible (BeFo, 2020) to legislation in the planning policy.Despite the introduction of 3D property rights, it is relevant to note that no amendments were made to the PBL following the introduction of 3D property formations (Boverket, 2004).Similarly, none were made to the Act on Technical Requirements on Construction Works (BVL), which guides technical requirements.Whilst the PBL remains applicable to subsurface space, as to any other property, the lack of specific provision or attention to the subsurface makes appropriate governance difficult.
Different codes can apply to specific underground functions such as transport types and provide specific rights, e.g., property rights (SGU, 2017) to railways, but not to roads, which are not considered their own property.However, it is unclear how planning and environmental legislation together apply to the underground and how these could support a holistic view (BeFo, 2020) of underground planning.The environmental code, for example, places requirements on groundwater diversions, which mostly become applicable on larger projects and require Land and Environmental Court permits (SGU, 2017).On the other hand, small underground projects can sometimes be carried out without being subject to the environmental code (SGU, 2017).Covered by the legislation on environmentally hazardous activities, heat pumps, for example, are a notifiable project and may require permission from the municipality (SGU, 2017).Legislation is not an abstract or singular tool, and Norrman et al. (2016) note that it should be borne in mind that legislation is complemented by political goals, policies and guidelines.

Addressing gaps and challenges for the future of underground urban planning in Stockholm
The projected influx of new residents to Sweden's capital (Stockholms stad, 2022b) will likely exert significant pressure for space and resources on the Stockholm municipality and its metropolitan region (Fig. 1).Moreover, the city's efforts towards low-carbon transition and potential impacts from climate change are likely to further strain its already limited capacity for management.Thus, Stockholm will inevitably need to explore the possibilities of utilising underground space to address these challenges and pressures in the future, as many cities globally are doing (for example, Singapore, which instigated an Underground Master Plan Task Force including benchmarking studies, see Zhou and Zhao, 2016;von der Tann et al., 2020).
Although city planners and engineers have, over the years, developed intricate subsurface infrastructures, the Stockholm metropolitan area is characterised by the "first comefirst serve" approach (Subsection 4.3) and currently lacks a long-term comprehensive plan or strategy for the utilisation of UUS (Subsection 4.4) (BeFo, 2020).Indeed, Stockholm lacks comprehensive planning tools that would recognise the underground as part of the larger city puzzle (BeFo, 2020).Below, we identify several distinct but interrelated gaps and challenges preventing the immediate integration of UUS into strategic decision-making for UUP in Stockholm.Of these, a significant (and interwoven) barrier to improved strategic planning is that the city's underground is not rendered visible in planning laws and policy (BeFo, 2020) and thus is of no use to the planners as well as being excluded from their tools.

Sectoral planning practices and short-term perspectives
Underground urban planning as a collaborative practice is not in place yet, and the utilisation of UUS is mostly driven by sector and guided by learning-by-doing instead.Learning-by-doing is a vital part of the process of developing effective UUP practices.Historically, major subsurface engineering projects in Stockholm have primarily aimed to alleviate traffic congestion between expanding suburbs and increasingly compact central areas of the city or have been legacies of Cold War defence planning (Nerlund, 2017).Consequently, the city's short-term needs take precedence over long-term perspectives, impeding the optimal utilisation of underground space in the future.
As is common globally (see Volchko et al., 2020), sectoral planning and expertise prevail.While there are certainly some examples of successful underground development projects in the city of Stockholm, as illustrated in Subsection 4.3, it is also important to work towards more structured and systematic approaches to underground planning to ensure the success and durability of these projects over the long term (Admiraal and Cornaro, 2020), as well as opening opportunities for synergies between the different types of utilisation of the subsurface (Admiraal, 2006).Moreover, SGU (2017) highlights that the absence of planning tools in Sweden that govern the general planning of the subsurface is perceived as a problem by stakeholders.This absence leads to subsurface planning occurring at a project level rather than a societal level.Moreover, multiple permit and legal processes can be required prior to projects commencing (SGU, 2017).The lack of clear and established planning practices presents a significant challenge for the development of UUS in Stockholm municipality, including the potential for conflicts with existing underground infrastructure due to the proliferation of ad hoc and inefficient underground developments.Additionally, as BeFo's report (BeFo, 2020) concludes, there is a need for systematic management of subsurface questions across the various governance scales in Sweden.We argue that Stockholm's learning-bydoing process should also include learning from other cities, such as neighbouring Helsinki (Vähäaho, 2014;Vähäaho, 2016), that have been successful in maturing their expertise in UUP to avoid limiting the potential benefits that UUS can bring to the city.Helsinki has a long history of inventorying its underground space, with allocation plans maintained since the 1980's.There the masterplanning process has focused on legal impacts such as prioritisation of development types and encouragement of consideration of the subsurface as an alternative to surface facilities (Vähäaho, 2016).In the application of the Deep City method to Geneva, Li et al. (2013) observe a successful integration of underground considerations into policy instruments, the development of information databases and improvements in the promotion of underground issues for surface revitalisation.This success is, however, built on an existing platform for knowledge sharing, as well as coordination with existing planning processes (Li et al., 2013).In the case of Stockholm, the existing knowledge sharing and collaboration are fragmented, and the existing planning processes are limited by scope.

Leadership (and vision)
Stockholm municipality lacks a clear sense of vision and leadership in anticipating or envisioning the demand for its subsurface resources and maximising the benefits of UUS, particularly in the long term.This is evident in how underground use is barely referred to in the city's governing documents, indicating its relatively low priority.For example, Stockholm's Urban Mobility Strategy briefly mentions the need for additional underground parking spaces to improve access to curbs on the ground (Stockholms stad, 2022a).The City Plan (Stockholms stad, 2018) refers to the underground in terms of potential space for expanding the Metro system and placing electricity cables below ground, exemplified by the Stockholms Ström project.As mentioned earlier, although Stockholm City has the mandate to govern land use on the municipal level, Region Stockholm is responsible for planning public transportation within the administrative borders of the county.Amid the new Metro construction being in full force and 18 new stations to be opened by 2035, policymakers in Region Stockholm envision that more needs to be done and plan to investigate at least five additional new Metro routes to be in place after 2040 (Beckman, 2023).Such sectoral long-term plans aimed at enhancing public transportation in the Stockholm region and its metropolitan area have the potential to come into conflict with other envisioned underground uses by the Stockholm municipality.
Central administration of an underground masterplan by the city planning department, as is the case in Helsinki (Vähäaho, 2016), would require a cross-sectoral and cross-scale collaboration, which Stockholm may find difficult in its current praxis.Without a clear sense of direction, Stockholm may miss opportunities to utilise its subsurface resources more effectively and sustainably, leading to wasted potential and environmental damage.Moreover, it is important to note that multidimensional conflicts can arise in the absence of a comprehensive and integrated strategy that connects various sectoral policies concerned with the city (sub)surface.Such conflicts may involve competing interests and priorities over different time scales, leading to challenges in achieving sustainable and inclusive UUP processes (Bartel and Janssen, 2016).Additionally, the lack of leadership in this area may result in inconsistent policies and regulations, hampering private sector investment in underground development and reducing the city's competitiveness.
To promote a lasting strategy for UUP that contributes to sustainable and resilient city development, it is crucial to involve key stakeholders, such as urban decision-makers, and demonstrate the potential benefits of an urban underground future.As Admiraal and Cornaro (2016a) aptly point out, convincing decision-makers is not limited to politicians, but it also involves those who hold formal and informal influence over project outcomes, including elected officials, public servants, the private sector and society at large.The challenge in developing underground urban spaces is that they are often not as visible or accessible as above-ground areas (Doyle et al., 2016).This can make it more difficult to generate interest and support for these projects and to identify potential uses and benefits (Volchko et al., 2020).Recognising the city's dynamic and multiple aspects and sharing this knowledge is important for forming new concepts of "Post-Carbon Cities" which can support urban leadership (Delmastro et al., 2016).

Information flows
Effective decision-making requires the appropriate supply of underground information (Mielby et al., 2017).There is limited access to and flow of geospatial information concerning the subsurface in Stockholm's metropolitan area, which may lead to increased project costs, delays, and safety risks and thus poses significant obstacles to the shortand especially long-term development and use of urban underground space.Moreover, the lack of such information makes it difficult for stakeholders to make informed decisions about underground development, hindering private sector investment and reducing the potential benefits for the city.Data is often collected by private consulting firms on behalf of the larger sectoral client agencies, and official responsibility for the collation of data is spread across multiple agencies depending on its type and expected use (Naturvårdsverket, 2000).BeFo notes that there are several ongoing projects across Sweden to try and create national collections of geo-data in digital formats.However, these are additionally hampered by issues around state security, which have impacted collection as well as access (BeFo, 2020).
Geospatial information, including geological data and detailed maps of underground infrastructures, is critical for effective planning and constructing underground projects (von der Tann et al., 2018).Therefore, there is a need to improve access to and flow of geospatial information concerning Stockholm municipality and its metropolitan area.Updating and maintaining detailed maps of underground infrastructure and geological data centrally can help planners and designers identify potential risks and opportunities in urban underground spaces.This can also enable stakeholders to make informed decisions about underground development, increasing private sector investment and the potential benefits for the city.Helsinki's masterplanning is founded upon a longterm collection of GIS information (Vähäaho, 2016), and the masterplan is now readily accessible through an online portal.Singapore, too, has highlighted GIS data access as a key road-block on the route to strategic subsurface planning, putting a single administrative body in charge of collecting and coordinating subsurface data (van Son et al., 2018).

Trade-offs and strategic decision-making
Demand for and trade-offs associated with developing urban underground space in Stockholm are poorly recognised (e.g., conflicts for urban underground space, climate change impacts).On the one hand, the growing population and urbanisation trends drive the need for more space in cities, including the underground.The development of underground spaces offers many benefits, such as increased livable space, reduced congestion, and improved resilience to natural disasters.On the other hand, developing underground spaces also involves trade-offs, such as conflicts with existing underground infrastructure, environmental impacts, and safety risks due to the geological complexities underground.For instance, a tunnel that hits an unanticipated permeable zone (e.g., a deformed layer or aquifer) may lead to the lowering of the water table and possibly soil subsidence if close to the surface.Additionally, underground spaces may be vulnerable to flooding and other climate change impacts, such as rising groundwater levels.Careful consideration around the utilisation of interacting resources is thus required (Cui et al., 2021).This requires the integration of tools and knowledge within urban planning expertise to enable effective decisionmaking (Hooimeijer and Tummers, 2017).Maring and Blauw's (2018) Asset Management approach to subsurface management aims to provide insight into relevant opportunities and risks to elucidate the complex decision-making field for stakeholders, but there remains an important aspect to be considered -the broadening of subsurface perspectives.
Assessments of multi-dimensional interactions are bound to the visions driving UUS development as well the tendencies of surface planning to see the subsurface as a void or merely a technical support layer (see, e.g., Lambertucci, 2019;Labbé, 2016;Melo Zurita, 2020).Despite progress in the literature on the subject, it is today still natural to think of the UUS in Stockholm mainly as a suitable place for accommodating infrastructure for transportation, i.e., a pancake stack with certain layers devoted to a certain type of infrastructure.A depth-based planning model developed by SGU's working group splits transport tunnel types by appropriate depth for a dense urban context (SGU, 2017).There is a risk, however, that not acknowledging a broader and fully 3D perspective (as advocated by the BeFo report) may hamper the development of the city.Allocating suitable underground space for other functions may increase the value of the UUS, and transportation should not only connect points A and B on the surface but also be planned for, e.g., recreational hubs or shopping areas that may well be placed underground (Cui and Lin, 2016).This will allow more space on the surface for living, preserving proximity to other vital functions at the surface (Huanqing et al., 2016).
In the drive to make subsurface interactions visible, Volchko et al. (2020) note that the perspective of multiple-resources is widely adopted.In this view, ecosystem services and Geosystem Services (GS) support man-made and natural assets, although there are ongoing discussions on the exact definitions of GS and their implications for its implementation (Frisk et al., 2022).GS perspectives have the potential to contribute to holistic planning, especially in facilitating communication across disciplines (Frisk et al., 2022).They can also remedy the underestimation of the value GS provide to sustainable development, but Van Ree et al. (2017) caution that existing valuations of certain subsurface resources are already embedded in societal processes (see also Van Ree and Van Beukering, 2016).Both SGU (2017) and Öberg and Sjöholm (2019) discuss the potential of GS in Swedish governance, observing its ability to help visualise conflicts between resources and their benefits to wider urban planning.Whilst exploratory application of GS is underway in Gothenburg (Norrman et al., 2021), aiming to promote recognition of complex interactions in the underground, the broader lack of holistic governance and knowledge sharing in Stockholm must be first addressed to enable complex decision-making.Additionally, BeFo recommends that the application of a GS perspective should be synchronised with data collection requirements to ensure appropriate information to support informed decision-making (BeFo, 2020).

Development drivers
Finally, the transformation of surface space in attractive locations from low-value to high-value buildings is a common trend in urban development.However, without including underground planning, the transformation may be suboptimal in the longer perspective (Bobylev, 2016).Developing surface space without considering the potential for underground development may limit the available space for future growth and development, as well as urban sustainability, resulting in missed opportunities and increased costs (Norrman et al., 2016).In addition, it may lead to inefficient land use and the proliferation of highdensity buildings that do not necessarily holistically reflect the needs of the community or promote sustainable development (Price et al., 2016).There is a need for increased attention on the interplay between sociopolitical, geophysical, and technical obstacles to implementing sustainable planning of the underground in a better manner.As is evident from existing literature, integrating subsurface management into urban planning at the city level is crucial.Hooimeijer and Tummers (2017) note that working towards sustainable cities is a key driver for integrating UUS into urban development processes but requires "primarily a change of culture".This requires coordinated policy changes and clear visions to drive changes in practices, particularly focusing on planning practice (Hooimeijer and Tummers, 2017).In Stockholm, envisioning future sustainable underground use may be challenging due to a wider reluctance to engage in discussions around conflicting sustainable development perspectives.Metzger (2013) observes a tendency to favour the "depoliticisation of environmental issues" in their discussion of increasingly technical approaches to urban planning.Our argument is that a focus on the multi-dimensional nature of the challenges and opportunities related to underground space development is required, emphasising the potential of planning tools like master plans to communicate guiding visions that integrate development goals.

Conclusions
We suggest that further research is necessary in several key areas to facilitate the effectiveness and sustainability of long-term urban underground use and planning in Stockholm municipality and its metropolitan area (Fig. 5).
First, it is necessary to identify and develop practices for engaging policymakers and other relevant stakeholders that can result in improved leadership and a more inclusive planning process.This would also entail creating educational tools to ensure that relevant stakeholders (e.g., government officials and city planners) have the knowledge and skills necessary to participate and act effectively in the planning process.Second, further advances are required to ensure timely access to and appropriate allocation of geospatial information about the different stages of the planning process (e.g., identifying which actors require access to what information, when they need it, and how it should be presented to them).Third, there is a need to study and develop decision support tools for managing potential trade-offs between competing priorities (e.g., costs and benefits) of various planning options, along with exploring financial challenges and fiscal models needed to encourage urban underground use (e.g., public-private partnerships).Consideration of how these tools complement or adapt existing local praxis is critical.Fourth, it is essential to critically consider societal and cultural values in the planning process, including understanding how different stakeholders value different aspects of urban life and how these values affect planning decisions.We stress that these avenues of further research are intrinsically co-dependant if long-term contributions to a sustainable city are to be realised.Attempts to implement better underground praxis should reflect on interrelations across this spectrum of current challenges, focusing on collaborations for a holistic approach to the planning of underground space.Integration into wider urban planning praxis requires dissolving the isolation of the underground as "other" and instead focusing on commonalities with wider urban development perspectives.
Finally, it is important to address the study's certain limitations that warrant attention.These limitations encompass the incorporation of GS and human-centred perspectives, which were beyond the current scope of our research.On the one hand, the appropriate allocation of geospatial information demands further investigation of the potential of a holistic GS lens in Stockholm City -i.e., using advanced mapping and data analysis techniques -to better understand the complex and intrinsic relationships between different surface and subsurface biotic and abiotic systems and how they interact (e.g., Bobylev et al., 2022;Frisk et al., 2022;Van Ree et al., 2017).The GS approach allows for identifying the societal challenges and drivers associated with (long-term) sustainable subsurface use, such as the increasing demand for space in urban areas.Van Ree and Van Beukering (2016) argue that the GS concept provides a systematic framework for sustainable underground use that not only informs but also guides policymakers and urban planners in making strategic decisions regarding subsurface in four areas: geosystems encompassing geological, geotechnical, structural, and geohydrological characteristics of the subsurface; services representing the valuable functions and benefits that the subsurface environment offers to society; the values associated with GS, comprising both their economic and noneconomic benefits; and governance processes that guide the use and protection of geosystem services, including policies, regulations, and strategies to ensure the sustainable utilisation of subsurface resources (Van Ree and Van Beukering, 2016).Stockholm's planners and policy makers could benefit from using a GS framework to better acknowledge and identify subsurface risks and opportunities (e.g., Finesso and Van Ree, 2022).Steps in this direction have already been made with the application of GS to a pilot study in Gothenburg, with findings suggesting it can support the sustainable use of subsurface resources (BeFo, 2020).The pilot demonstrated GS's benefits, such as communicating the value of the underground and providing a sound geoscience base to further dialogue on utilisation.The pilot's proposed SUB-matrix model, adapted to local municipal planning specificities, describes geosystem services in detail and is applied at a project scale within Gothenburg.It has potential to help strategically plan subsurface use (long-term) supported by "analysis of possible conflicts and synergies", including optimising the use of underground spaces for infrastructure, transportation, and utilities in relation to natural supporting services (Norrman et al., 2021).Yet, BeFo (2020) also notes that for GS to effectively support planning decisions, it must be integrated with surface praxis and that in Sweden "the formal physical planning and approval buildings in the subsurface needs to be clarified and developed".
On the other hand, a broader recognition of the potential value of underground space and incorporation into Stockholm's development visions is related to the human experiences of underground spaces, a subject that merits further research (e.g., Lee et al., 2016;Lee et al., 2017;Soh et al., 2016).Perceptions of the underground as a primitive or unsafe space of darkness and containment can negatively impact user experiences and prevent greater uptake of use (e.g., Mohirta, 2012).Therefore, as Soh et al. (2016) demonstrate, the importance of enhancing social acceptance of underground spaces becomes evident when one considers the holistic improvement of subsurface use with a focus on human psychology, health and safety.This multifaceted objective underscores the need for a comprehensive approach to transforming underground environments into spaces that are not only functional but also conducive to the physical and mental well-being of individuals (Soh et al., 2016).For example, the type and level of lighting and the architectural characteristics of underground space, as well as its connectivity with the above ground, are important for well-being, especially if longer-term habitation or working in the subsurface is to be expected (e.g., Kaplan, 1993;Roberts et al., 2016;Zhao and Künzli, 2016).Moreover, in a more extensively vertical urbanism, a better understanding of dominant cultural imagery at local levels could influence the design of underground spaces, supporting the diversity of subterranean spaces and identities (Lee et al., 2016).Such themes around occupant perception and human-centred design will be important when considering long-term support for Stockholm's UUS use across the public, expert and political governance levels.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Fig. 2 .
Fig. 2. Bedrock map of Stockholm municipality showing deformations.Data sourced from the Swedish Geological Survey (SGU) and Swedish Land Survey (Lantmäteriet).

Fig. 3 .
Fig. 3. Elevation map of Stockholm Municipality with deformations, exposed bedrock areas and simplified quaternary deposits at 6-15 m depth range.Data sourced from the Swedish Geological Survey (SGU) and Swedish Land Survey (Lantmäteriet).

Fig. 4 .
Fig. 4. Examples of existing and planned underground infrastructures in Stockholm.Data sourced from the Swedish Geological Survey (SGU), Swedish Land Survey (Lantmäteriet) and Esri base maps.

Fig. 5 .
Fig. 5. Visualisation of key issues and recommended action areas for Stockholm's subsurface praxis.

Table 1
Summary of planning in Sweden by sector.