Functionally Fractal Urban Networks: Geospatial Co-location and Homogeneity of Infrastructure

Christopher Klinkhamer, Elisabeth Krueger, Xianyuan Zhan, Frank Blumensaat, Satish V. Ukkusuri, P. Suresh C. RaoarXiv 2017 · DOI · arXiv

Abstract

Just as natural river networks are known to be globally self-similar, recent research has shown that human-built urban networks, such as road networks, are also functionally self-similar, and have fractal topology with power-law node-degree distributions (p(k) = a k). Here we show, for the first time, that other urban infrastructure networks (sanitary and storm-water sewers), which sustain flows of critical services for urban citizens, also show scale-free functional topologies. For roads and drainage networks, we compared functional topological metrics, derived from high-resolution data (70,000 nodes) for a large US city providing services to about 900,000 citizens over an area of about 1,000 km2. For the whole city and for different sized subnets, we also examined these networks in terms of geospatial co-location (roads and sewers). Our analyses reveal functional topological homogeneity among all the subnets within the city, in spite of differences in several urban attributes. The functional topologies of all subnets of both infrastructure types resemble power-law distributions, with tails becoming increasingly power-law as the subnet area increases. Our findings hold implications for assessing the vulnerability of these critical infrastructure networks to cascading shocks based on spatial interdependency, and for improved design and maintenance of urban infrastructure networks.

Introduction

The structural organization of natural systems, such as river basins and vascular networks, has long been recognized as fractal—exhibiting self-similarity across multiple scales. Recent urban science has extended this observation to human-made road networks. However, the "hidden" infrastructure that sustains modern life—specifically sanitary and storm-water sewer systems—has remained largely unexamined through the lens of complex network topology. This research investigates whether these subterranean networks follow the same functional fractal patterns as their surface-level counterparts and how their physical co-location influences urban resilience.

Methodology

To analyze the functional rather than just physical connectivity of urban systems, the study employed Hierarchical Intersection Continuity Negotiation (HICN). Unlike "primal" graphs where intersections are nodes and streets are edges, HICN uses a "dual" representation. In this model, individual street or pipe segments are treated as nodes, and their intersections are the edges. This approach better reflects the operational flow of resources and services through the network.

The researchers analyzed high-resolution GIS data from a major US city, encompassing approximately 70,000 nodes over a 1,000 km² area. To test for scale-invariance, the team extracted 25 random points across the city and generated nested circular subnets with radii ranging from 1.25 km to 7.5 km. Statistical validity was ensured through double power-law fitting, which accounts for the "finite-size effects" inherent in city-scale data.

Functional Homogeneity and Scale-Free Topology

A primary finding of the study is the existence of "functional homogeneity" across the urban landscape. Once a subnet reaches a critical threshold—approximately 20 km² in area or a population of 50,000—its topological properties become statistically consistent. Whether in the dense urban core or the sprawling suburbs, the underlying "DNA" of the infrastructure remains the same.

Both road and sewer networks demonstrated scale-free functional topologies, characterized by power-law node-degree distributions. The power-law exponents ($\gamma$) were found to be approximately 2.5 for roads and 2.9 for sewers. These values are consistent with other complex "small-world" networks, suggesting that urban infrastructure evolves toward a mathematically optimized state of connectivity.

Structural Divergence: Loops vs. Trees

While both systems are fractal, they serve different functional purposes, reflected in their topology. Road networks exhibit a high average node-degree (~4), indicating a highly looped structure. This redundancy is essential for urban transit, allowing for multiple routes between any two points.

In contrast, sewer networks exhibit a lower node-degree (~2), resulting in a tree-like structure. This topology is optimized for the gravity-driven collection of waste, channeling flows from many disparate points toward a single treatment or discharge location. While efficient for waste management, this tree-like structure is inherently more vulnerable to single-point failures than the redundant road network.

Geospatial Co-location and Strategic Aversion

The study also quantified the physical relationship between these networks. Approximately 66% of the total sewer length is co-located within road rights-of-way. However, the researchers discovered a phenomenon termed "strategic aversion." While the networks are physically close, the most critical "hubs" (nodes with high centrality) in the road network are rarely located directly above the most critical sewer mains.

This spatial decoupling suggests an evolutionary or planned resilience strategy. By avoiding the concentration of high-importance infrastructure in a single geospatial footprint, the city reduces the risk of cascading failures. A single catastrophic event, such as a major sinkhole or pipe burst, is less likely to simultaneously disable both a primary transit artery and a critical waste-water trunk.

Conclusion

This research confirms that urban sewer networks, like roads and rivers, are functionally fractal. The discovery of topological homogeneity and strategic aversion provides a new framework for urban planning. By understanding these mathematical signatures, engineers can better assess the vulnerability of critical infrastructure to cascading shocks and design more resilient systems, even in data-poor environments. Future work should expand this analysis to other utilities, such as power and telecommunications, to develop a truly holistic model of urban interdependency.

Summary updated

Abstract

Just as natural river networks are known to be globally self-similar, recent research has shown that human-built urban networks, such as road networks, are also functionally self-similar, and have fractal topology with power-law node-degree distributions (p(k) = a k). Here we show, for the first time, that other urban infrastructure networks (sanitary and storm-water sewers), which sustain flows of critical services for urban citizens, also show scale-free functional topologies. For roads and drainage networks, we compared functional topological metrics, derived from high-resolution data (70,000 nodes) for a large US city providing services to about 900,000 citizens over an area of about 1,000 km2. For the whole city and for different sized subnets, we also examined these networks in terms of geospatial co-location (roads and sewers). Our analyses reveal functional topological homogeneity among all the subnets within the city, in spite of differences in several urban attributes. The functional topologies of all subnets of both infrastructure types resemble power-law distributions, with tails becoming increasingly power-law as the subnet area increases. Our findings hold implications for assessing the vulnerability of these critical infrastructure networks to cascading shocks based on spatial interdependency, and for improved design and maintenance of urban infrastructure networks.
BibTeX
@misc{klinkhamer2017functionally,
  author = {Christopher Klinkhamer and Elisabeth Krueger and Xianyuan Zhan and Frank Blumensaat and Satish V. Ukkusuri and P. Suresh C. Rao},
  title = {{Functionally Fractal Urban Networks: Geospatial Co-location and Homogeneity of Infrastructure}},
  year = {2017},
  eprint = {1712.03883},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1712.03883},
}
Figure