Hacker News Reader

Comment by jorleif | Hacker News Reader

These kinds of techniques are often claimed to be effective against flash floods, which makes sense since they slow down water transfer to rivers. Does anyone know any studies about how effective it could be against seriously heavy precipitation of the scale of Harvey or South Asia floods right now? In other words, how well does the approach scale?

View on HN

Civil engineering has pretty well quantified the rate at which water is absorbed by certain soils, accounting for groundcover, trees, pavement, local detainment, etc. Berlin's sponge mission is baiscially distributed detainment, working to offset hardscaping (driveways and roofing). In other words, it's a well known and very predictable science, but we (we being US land development industry) has not cared. HOWEVER rain at the scale of Harvey cannot practically be planned for beyond "don't build on land that's lower than average"

Well, that's partly true. A lot of the flooding is due to the fact that the entirety of Texas doesn't care about hydrology, and all that water is ending up downstream, in Houston.

There'd still be flooding, and it would still be bad, but a certain degree of flooding was avoidable.

Deregulation at work

The other thought, I imagine, is that they felt the probability was low that they would see a flood of this scale even within their lifetime; that ended up being a poor bet. Even so, flood insurance is also cheaper when the perceived likelihood of a flood is low.

Since cities usually survive much more than one human lifetime, this is not good reasoning.

People build houses, cities don't.

As you almost certainly know, the city controls permissions for every building.

A city that gives planning permission for a development that is probably going to be destroyed with loss of life in the next hundred years is not serving people well.

Eh it is wholly true. Distributed detention schemes still have a finite volume of containment, and once the detention zones are saturated, those areas behave equally to pavement for each additional unit of water. 50 inches of water across the majority of Texas with severe storm surge cannot be practically prepared for beyond insurance, evacuation, and/or a boat. To say “it was avoidable” is a strong and expensive assertion requires some hard backup. Heres an introduction to runoff coefficient (disclaimer! exciting stuff!). Everything is based on a “design storm” with a specified duration and intensity, where those two variables lead to a sigmoid curve (as i recall) for runoff, or the greater the intensity and longer the duration, the less water that gets detained. http://www.brighthubengineering.com/hydraulics-civil-enginee...

I remember a demonstration from my childhood where water is released into 2 similar models. One model is dry, one model is wet. The water quickly overtakes the dry model and splashes on the other side (as in a flash flood). On the wet model, water takes longer to make it to the other side. Sort of how a wet sponge is far more absorbent than a dry sponge.

I know I didn't really answer your question, but these models were made to represent fields and swamps. So yes, I believe there are lots of studies on large scale effects.

I don't know of any studies but I would imagine a lot of variables come into play.

Soil Type & how deep of soil effects how much water the ground can soak up & how fast the water will run. All would be slower runoff than concrete.

Also the slope of the land would make a difference.

These are just the obvious ones. Bottlenecks, whether created on purpose or not would have a huge impact. Where I live we have large snowfall in the winter which creates ice jams & flash flooding in the spring.

I personally like the idea of this.

Every inch of rain absorbed makes a rather dramatic difference. Because, floods are moving a lot of water but they also contain a lot of water which is time shifted over a longer period of time.

Think of it like a building being evacuated though a stairwell. Being say 10 vs 11 floors can make a significant difference in average wait time. Except average weight time is equivalent to the height of flood waters.

Reply on news.ycombinator.com