Integrated Hydrologic–Hydraulic Framework for Culvert Siting at Road–Waterway Intersections in Ungauged Basins

Imagine the Earth as a giant, living sponge. When it rains, this sponge soaks up some water, but if the rain comes too fast or too hard, the excess has to go somewhere. It rushes down hills, gathers in streams, and eventually wants to flow into rivers. Now, imagine we build a road right across this path. The road acts like a wall, blocking the water's natural journey. If we don't give the water a way to pass underneath, it will crash over the road, washing it away and leaving cars stranded. This is the problem of "culverts"—those big, sturdy tunnels we build under roads to let water through.

Designing these tunnels is tricky, especially in places where we don't have a long history of measuring exactly how much rain falls or how fast rivers flow. These are called "ungauged basins," or in simpler terms, "blind spots" where engineers have to guess the rules of the game. To solve this, scientists use computer models that act like digital sandboxes. They simulate the rain, the rolling hills, and the rushing water to see how a flood might behave before they ever dig a single hole in the ground. This paper is about building a better, more connected way to play in that digital sandbox, specifically for a region where the terrain is rugged and the data is scarce, ensuring that when the real storm hits, the road stays safe.

The Story of the Digital Water Tunnel

In the Kurdistan Region of Iraq, a new highway is being carved through the hills of New Zakho City. It's a beautiful project, but it cuts right across several natural drainage channels where water rushes down during heavy rains. The engineers faced a tough puzzle: How do you build the right size of culvert (the tunnel under the road) when you don't have a record of past floods to tell you how big the water might get?

The team, led by researchers from universities in Iraq and Sweden, decided to build a super-connected "digital twin" of the area. Instead of using three different tools that barely talk to each other, they linked them up like a relay race team passing a baton perfectly. First, they used a tool called WMS to map out the land, figuring out exactly where the water would gather and how fast it would run based on the soil and the shape of the hills. Next, they handed that data to HEC-RAS 2D, a powerful simulator that acts like a high-tech video game engine. This tool didn't just guess the water level; it actually "played out" the storm, showing exactly how deep the water would get and how fast it would be moving across the entire landscape for two different types of storms: a big one that happens once every 50 years, and a monster one that happens once every 100 years. Finally, they fed those results into Bentley Culvert Master, a specialized calculator that designs the actual tunnel dimensions.

What They Found in the Simulation

The team started by looking at 22 years of rain records from a local weather station. They found that the biggest 24-hour rainstorms in this area can drop about 104 mm of rain in a 50-year event and 114 mm in a 100-year event. That's not a huge difference in rainfall amount, but the water behaves very differently depending on the size of the hill it's coming from.

They identified four spots where the road crosses a stream. The smallest area feeding water to a crossing was just 0.99 km², while the largest was a massive 15.71 km². When they ran their simulations, the results were eye-opening. The smallest spot needed to handle a flow of 13.1 m³/s (cubic meters per second), but the largest spot needed to handle a roaring 111 m³/s.

Here is the big surprise: You might think that if the area is 16 times bigger, the water flow would be 16 times bigger. But in the real world, it's not that simple. The largest area actually produced a flow that was only about 6.4 times bigger than the smallest one. Why? Because the bigger hills had steeper slopes and different soil that didn't soak up water as well, creating a "runaway" effect where the water rushed faster and harder. This taught the team that you can't just use a simple math formula to guess the size of a culvert based on the size of the land; you have to look at the specific details of each spot.

Designing the Tunnels

Using their simulation data, the team designed the culverts. They found that for the smallest crossing, a twin-tube box culvert measuring 2.13 m × 1.22 m was perfect. It let the water through without backing up too much. However, for the two middle-sized crossings, the water was so fast (over 3.0 m/s) that it would have eroded the ground at the exit. To fix this, they designed the tunnels to be larger (triple-cell boxes) and added a special "rip-rap apron"—a layer of heavy rocks—to act like a shock absorber, slowing the water down so it wouldn't wash the road away.

The biggest crossing, with its massive 111 m³/s flow, needed a giant triple-cell tunnel measuring 3.66 m × 3.05 m. Even with this huge size, the water was still moving at 3.9 m/s, so they had to specify an even thicker and longer layer of rocks to protect the ground.

Why This Matters

The most important thing this paper shows is that you don't need a long history of flood records to build safe roads. By using free, open-source maps and connecting their computer tools tightly together, the team created a reliable blueprint for the culverts. They proved that even in a place where data is scarce, you can design infrastructure that stands up to a 100-year storm.

They also showed that the old way of doing things—just guessing or using simple rules—might not work in these complex, hilly areas. Their "digital sandbox" approach, which simulates the water's speed and depth in detail, gives engineers a much clearer picture of what to build. While they couldn't test this in a real flood (since they don't have historical data to compare against), the simulations were checked against standard engineering rules and matched up perfectly, giving them high confidence in their designs.

In the end, this framework is like a universal translator for engineers. It takes the messy, complex reality of a mountainous, data-scarce region and turns it into a clear, safe plan for keeping roads open and people safe when the rain comes.