Should underground infrastucture location data be included in digital twins ?

There are fundamental differences between above- and below-ground data which may limit the usefulness of underground data in digital twins. For location data about the underground there are often data quality issues, legal restrictions on sharing privately held data, and potential liabilities that can place limits on how this data can be used.

At the GEOBIM 2020 conference many of the presentations discussed developing digital twins of cities (Rotterdam, Helsinki, Pilsen, Athens), regions (Flanders), and even nations (Estonia, Singapore, UK). Many cities have 3D models of their above-ground infrastructure including buildings, transportation networks, parks, and other infrastructure typically captured in 2D or 3D imagery from overflights or in 2D GIS maps. But underground infrastructure is often neglected, even though it is recognized that water and wastewater, energy (gas, electric power, and district heating), and communications (fibre and copper) networks provide the life blood of the city.

It may be simply out-of-sight out-of-mind, but there is also a fundamental difference in what jurisidictions are able to do with representations of above and below ground infrastructure. Above ground, whatever can be photographed or scanned in the public space; from the street, from an airplane or drone, or from a satellite, in 2D or 3D, is generally unrestricted in application. It is necessary to remove recognizable people and vehicles for privacy reasons and there may be national security restrictions, but in general this data can be used without restruction to create an open, publicly available digital model or digital twin. The accuracy of the above-ground data in digital twins is generally high – for digitally scanned data accuracy even reaches the mm level. And because this is open data, provided “as is”, liability does not appear to have become an issue.

Underground the situation is different. First of all, privately owned property is generally off limits. For example, pipes and cables running on private property and connecting homes and business to the municipal water, gas, power, and telecom services are normally missing in digital city models. Secondly, in most jurisdictions some or all of the utility and telecom infrastructure in the public right-of-way is privately owned. There is a range in how much of the infrastructure in the public right-of-way is publicly owned and how much is private. Many cities own water and sewer services and this data is often open and publicly accessible. But electric power, gas, and telecom are typically privately owned. Private ownership, especially in the case of a competitive market such as telecom, imposes a fundamental restriction on public access to this data in the context of an open digital twin.

For example, the City of Rotterdam is proud that its digital twin is open and accessible to the public and that it includes above and below ground infrastructure. But, as Roland van der Heijden explained at GEOBIM, while about 2/3 of the underground utility data is open and accessible, there are restrictions on the other 1/3 because these represent privately owned facilities.

Helsinki is one of the early adopters of the digital twin concept and actually has two models, one derived from imagery and intended for visualization applications. The other is a semantic digital twin based on the CityGML data model which has extensions for utilities. In his GEOBIM talk Jarmo Suomisto made it clear that the city has no plans to include location data about underground infrastructure. Jarmo gave several reasons for this decision all of which are important issues that in general apply in any municipal, regional or national jurisdiction. In Helsinki location data for underground utilities is owned and maintained by private utility and telecom owners. Access to this data requires a data sharing agreement between the city and the network operators. Furthermore, there are questions about whether the quality of the utility location data (completeness, accuracy and currency) is up the city’s standard for its digital twin. Finally, there is the major issue of liabilities. Who is responsible if someone, say a construction contractor, uses this data and hits an underground utility, the city, the network operator or the contractor ?

In Estonia, the situation is different. National legislation requires that all utility network owners with infrastructure in the public right of way or under land publicly owned provide their data to the national registry. The legislation does not apply to small private owners (private houses on private land). Furthermore, when changes are made to the network, either installing new infrastructure or renovating old infrastructure, the data provided to the national registry needs to be refreshed. Even more interesting, when during excavation it is discovered that the location of network infrastructure is incorrect, it is required that the registry be updated with the corrected information.

Slovenia has a process with similarities to the Estonian system. Network operators are required to submit location data for underground infrastructure to the Geodetic Administration of the Republic of Slovenia (GURS) which maintains open, publicly accessible databases of underground utility and telecom infrastructure, parcel files and other data. In the case of telecom network operators are subject to civil penalties if it is found that in some area they have not provided records of the location of their underground cables. For this and other reasons the quality of the underground data has been improving and it is expected that within a couple of years, excavators will be able to rely on the public database instead of having to contact network operators directly for records of the infrastructure in the area of a planned excavation.

An important question for any data included in a digital twin is fitness-for-purpose. The most common use case for city digital twins is planning. Other common use cases include noise abatement, wind and heat control, zoning compliance, permitting and right-to-light. Each use case has specific data requirements. Many don’t require survey-grade accuracy in location, it may only be necessary to refresh the data once or twice a year, and the level of detail may be low. For underground utilities the OGC MUDDI underground standards project has identified several uses cases for information about the location of underground infrastructure; routine street excavations, emergency response, utility maintenance programs, large scale construction projects, disaster planning and response, and smart cities programs. Each has different requirements for access, data quality, currency and level of detail.

The only instance I know of in North America of a province or state wide database that includes maps of most utility infrastructure is the Integrated Cadastral Information Society (ICIS) in British Columbia. ICIS maintains a database of geospatial data including cadastral and parcel map for the province, civic address registry, assessment data, utility infrastructure data contributed and maintained by major utility organizations operating in BC, and local government infrastructure data, including sewer, storm and water mains. ICIS is a non-governmental and non-profit organization that is supported entirely by its members including telecommunications and utility companies, local governments, First Nations, the Provincial Government, Federal Government departments, and private surveying companies.ICIS data is used for efficient communication between utilities, safe operations and emergency response, optimized infrastructure planning and operational service delivery, and shared spatial reference data for engineering and survey work. Data contributed by members may be refreshed one or twice a year. But the requirements of one call legislation are very specific and damage prevention during excavation is not a use case directly supported by ICIS. And this is even though British Columbia is the only jurisdiction in North America that I know of where network operators can opt to provide plans (paper or digital) of their underground services in response to one call requests. In fact, the last time I checked there was no relationship between ICIS and British Columbia One Call (BC1C).

The key question for jurisdictions developing digital twins and considering including the underground is for which use cases is underground infrastructure data with its current limitations – restricted access and incomplete, inaccurate and out-of-date data – fit-for-purpose ? For Helsinki it appears that there are no compelling use cases for underground data in its current state. While for other jurisdictions such as Rotterdam, Estonia and Slovenia there are sufficiently compelling reasons for including underground data in a digital twin.

This information was first published on https://geospatial.blogs.com/geospatial/2020/12/should-underground-infrastucture-location-data-be-included-in-digital-twins.html