Station Geography Precedes Station Ranking: Multiscale Territories and Spatial Policy Diagnosis in Taipei Metro

When we look at a city's subway system, we often see it as a simple map of lines and dots. But for the people who live there, a station is not just a dot; it is a gateway to a neighborhood, a hub of daily life, and a place where thousands of people cross paths every day. Urban planners and geographers have long tried to measure how well these stations work, asking questions like: Is the area around the station easy to walk through? Does it have enough shops and services? Is it well-connected to the rest of the city? For years, the standard way to answer these questions was to draw a circle around a station and count everything inside it. The idea was that if you walked a certain distance, say eight hundred meters, you were in that station's territory. This approach assumed that the world inside the circle belonged to that station alone, and that the world outside did not matter.

However, cities are messy, and people do not live in neat circles. A single subway station might have dozens of entrances scattered across a wide area, some on one side of a busy road and others on the opposite side. When you draw a circle around such a station, it often overlaps heavily with the circles of its neighbors. This means the same street, the same park, and the same group of people might be counted as part of two or three different stations at once. This creates a puzzle: if we are trying to judge which station is "better," what does a score actually mean if the territory it measures is shared with everyone else? This uncertainty is at the heart of a new study looking at the Taipei Metro, a massive network serving millions of passengers. The researchers wanted to know if we can truly rank stations against each other when their physical worlds are so deeply intertwined, and whether the way we draw our maps changes the story we tell about the city.

The study, led by Chia-Nung Li, introduces a new way of thinking called "Station Geography." Instead of treating a station as an isolated point, this approach looks at the station as a relationship between its position in the network, the physical space around it, and the people who use it. The team focused on twelve specific stations in Taipei, ranging from the massive central hub at Taipei Main Station to quieter stops on the edge of the city. They gathered a vast amount of data, including the exact locations of 388 different entrances, the number of people entering and exiting, and the layout of the six lines that make up the network. They then built a detailed assessment system using eighteen different factors, such as how friendly the sidewalks are, how many public facilities are nearby, and how well the station connects to the rest of the transit system.

The first major discovery was about the size of the territory. The researchers tested three different walking distances: 400 meters, 800 meters, and 1,200 meters. At the shortest distance of 400 meters, the areas around the stations barely touched each other. But as they expanded the radius to 800 meters, the situation changed dramatically. In the full network of 118 stations, the median amount of territory shared with a neighbor jumped to nearly two-thirds. By the time they reached 1,200 meters, almost every station's territory was overlapping with others to a degree of 94 percent. This means that at the scales where most people actually walk to catch a train, the idea of a station having its own exclusive neighborhood is largely an illusion. The stations share corridors and districts so thoroughly that counting the same population or facility for multiple stations is not a mistake; it is a reality of the city's layout.

Because of this overlap, the researchers found that the type of station matters more than a simple score. They used a method to group the stations based on their actual role in the city. The clearest split they found was between "core" stations and "local" stations. The core stations are the heavy hitters: they are major transfer points with many lines, huge passenger volumes, and high centrality in the network. The local stations serve specific neighborhoods, often with fewer entrances and less traffic. Within the core group, there are even finer distinctions, such as "transfer anchors" that act as massive hubs, versus "urban destinations" that serve commercial centers. This classification suggests that comparing a massive interchange station directly with a quiet neighborhood stop is like comparing apples and oranges; they are built for different purposes and serve different parts of the city's life.

When the team looked at how their complex scoring system worked, they found that the scores were heavily influenced by a station's position in the network. Stations that were central, had high passenger demand, and were close to the main hub tended to get higher scores. This makes sense, but it also means that a high score does not necessarily mean the local walking conditions are perfect. A station can be a vital, busy hub with a high score even if the sidewalks nearby are difficult to navigate. Conversely, a station in a quieter area might have excellent walking conditions but a lower score simply because it is not a major network hub. The study showed that the scores captured a mix of network importance and local quality, but they could not easily separate the two. This is a crucial distinction for planners: a high score might tell you a station is important, but it does not automatically tell you that the local environment is good.

The researchers also tested how stable these rankings were when they changed the assumptions about the data. They ran thousands of simulations to see if the order of the stations would hold up if the scores were slightly different or if the way they were calculated changed. The results showed that the very top station, Taipei Main Station, was a stable leader; it remained at the top no matter how the numbers were tweaked. However, the middle of the list was much more fluid. Stations that were ranked second, third, or fourth in one scenario might swap places in another. This suggests that while we can be confident about who the leaders are, trying to create a precise, numbered league table for the middle stations is misleading. The differences between them are often too small to be certain, and the order depends heavily on the specific rules used for the calculation.

To make sense of this complexity, the study moved away from a single ranking list and instead created four different "profiles" for each station. These profiles looked at different aspects of the station's life: how much pressure it faces from the sheer number of people, how complex its entrances and connections are, how well it is integrated into the wider city, and how active the local neighborhood is. This multidimensional view revealed that stations have different needs. For example, Taipei Main Station and Ximen Station were identified as having high pressure from passenger flow and complex interfaces, suggesting their main challenges are managing the crowds and the connections between lines. In contrast, stations like Tamsui, located on the edge of the network, faced different issues related to how well they connect to the rest of the city. This approach allows planners to see that a station does not have a single "problem" but a unique combination of challenges that require specific solutions.

The study concludes that the way we measure a station must match the reality of the space it occupies. Drawing a simple circle around a station and counting what is inside is no longer sufficient when those circles overlap so completely. Instead, we need to acknowledge that stations share their territories and that their performance is tied to their role in the larger network. A score is not a universal truth; it is a snapshot that depends on the boundaries we draw and the stations we compare it against. By separating the network importance from the local conditions and accepting that some rankings are uncertain, we can get a clearer picture of what each station actually needs. This does not mean we stop trying to improve our cities, but it does mean we stop pretending that a single number can tell the whole story. The real work lies in understanding the specific geography of each place, the people who move through it, and the unique challenges that arise when the city's spaces overlap.