A Directional and Sequential Viewshed Framework for Analysing Terrain-Controlled Visibility along an Escarpment Road

Imagine driving along a road carved into the side of a massive cliff. As you move, the world around you does not stay the same. Sometimes the road curves, and the view opens up to reveal a vast valley stretching out for miles. A moment later, the road dips into a cut through the rock, and the landscape disappears, replaced by a wall of earth and a narrow strip of sky. For a long time, scientists and engineers have studied how much of the world a person can see from a specific spot on a map. They use computer models to draw lines from a standing point to the horizon, calculating what is visible and what is hidden. This has been useful for planning where to build towers or parks. But this traditional way of looking at visibility treats the world as if you are standing still. It does not account for the fact that when you are driving, you are constantly moving, and the view changes with every meter you travel. It also rarely considers that the view might look completely different if you are driving in the opposite direction.

A team of researchers from Beijing Jiaotong University decided to look at this problem in a place where the terrain is dramatic and the road is constrained by steep slopes: the Great Rift Valley in Kenya. They focused on a specific 1.5-kilometer stretch of the road between Mai Mahiu and Nairobi. This section of the road is a classic example of an escarpment route, where the highway climbs or descends a steep edge of the earth. The researchers wanted to understand not just how much land is visible at any single point, but how that visibility changes as you drive, and how it differs depending on whether you are heading up the hill or down it. They built a new way of using computer mapping tools to track these changes, treating the road not as a series of static stops, but as a continuous journey where the view is constantly being revealed and hidden by the shape of the land.

To do this, the team placed sixteen imaginary observation points along the center of the road, spaced every 100 meters. At each point, they used a high-resolution digital map of the terrain to calculate exactly what a driver would see. They simulated a driver's eye level and drew lines of sight in every direction within a 200-meter radius, ignoring trees or buildings to focus purely on the shape of the ground. They then split the view at each point into two directions: the view looking uphill toward Nairobi and the view looking downhill toward Mai Mahiu. By comparing these two views, they could measure how much the landscape changed depending on which way the car was facing. They also measured the difference in visibility between one point and the next to see how abruptly the view opened up or closed in as the car moved forward.

The results showed that the road is a place of extreme visual contrast. The amount of visible ground varied wildly from point to point. At some locations, drivers could see nearly 82 percent of the surrounding terrain, offering a sweeping panorama of the Rift Valley. At other spots, visibility dropped to less than 10 percent, where the road was tightly enclosed by steep cut slopes and the view was almost entirely blocked. The average visibility across the whole stretch was about 40 percent, but this average hid the fact that the experience was anything but uniform. The researchers found that the shape of the land, specifically the steepness of the slope, was the strongest factor determining how much could be seen. Steeper areas tended to offer more open views, while gentler slopes often resulted in more enclosed, restricted views.

Perhaps the most surprising finding was how different the experience was depending on the direction of travel. The road is not a neutral corridor where the view is the same for everyone. When driving uphill, the average visibility was about 44 percent, but when driving downhill, it dropped to about 35 percent. More importantly, the pattern of what was visible changed completely. At one specific spot, a driver heading uphill might see a wide-open valley, while a driver heading downhill at the exact same location would see almost nothing but the road ahead and the wall of the cut slope. In other places, the reverse was true. This means that two drivers passing each other on this road are experiencing two entirely different visual worlds, even though they are on the same strip of asphalt. The researchers calculated a score to measure this difference, and found that for many parts of the road, the imbalance between the uphill and downhill views was significant.

The journey along this road is also defined by sudden shifts. The view does not change gradually; it often snaps from one state to another. The team identified specific moments where the visibility jumped or dropped by more than 20 percent over a very short distance. These are the points where a driver might suddenly emerge from a tight, enclosed tunnel of rock into a wide-open vista, or vice versa. These abrupt transitions create a rhythmic pattern of exposure and enclosure. Some sections of the road are consistently open, while others are consistently closed, but the most dynamic parts are where these states switch rapidly. The researchers found that these sudden changes happen more frequently and with greater intensity when driving downhill than when driving uphill, making the descent a more volatile visual experience.

This study suggests that understanding road safety and the driver's experience requires looking beyond simple maps of what is visible from a single spot. The shape of the land creates a dynamic visual environment that changes with every step of the journey and depends entirely on which way you are facing. For road planners and safety engineers, this means that a road segment cannot be treated as a single unit with one level of visibility. Instead, they need to consider the specific direction of travel and the sudden transitions that occur along the route. A section that looks safe for an uphill driver might be dangerously blind for a downhill driver. By recognizing these directional and sequential differences, authorities can better identify where to place warning signs, improve road markings, or manage speed to help drivers anticipate the changing landscape. The work provides a new way to see the road, not just as a path through the terrain, but as a sequence of visual experiences that are shaped by the earth itself.