The Risks of Hazard

What? Flood models are biased? 

Today is a big day for Intermap and for InsitePro — we are releasing FloodScope™ USA, the only flood model available that is built on a contiguous, manually-edited digital terrain model (DTM) for the lower 48 states and Hawaii.

As I've discussed before, elevation data is a key input into any flood model, so the ability to leverage a high quality DTM at full resolution is a significant achievement. Starting today, insurers can use flood zone determinations as a part of their analytics to better understand the risk of river flooding.

One of the most devastating forms of flood risk is coastal storm surge. The flooding is most intense at the shoreline, but surge flooding can affect locations 40 miles (65 km) from the coast. The amount of storm surge is dependent upon the size of the storm creating the surge, and the tide at the time of the surge. As building codes have improved, most damage caused by hurricanes is now due to the flooding associated with them — not wind damage. Some of the most devastating and expensive natural disasters in the USA have been largely due to storm surge, including Sandy and Katrina.

Flood modeling is entirely dependent upon elevation measurements, and a hidden but crucial aspect of any elevation measurement is the datum on which it’s based. The datum defines the height of "zero." Once again I've asked my colleague at Intermap, Mike Wollersheim, to contribute his knowledge to the blog. When it comes to explaining what’s up with datums, he’s definitely up to the task.

If I told you that your city has an elevation of 3,400 feet, most people assume that I am referring to the height above mean sea level. What else are you going to measure your heights from? It seems obvious unless you really think about it. Then you start to ask questions like:

“If I am in the mountains 500 miles from the coast, how did anyone ever figure out what mean sea level is right here?” 

The use of risk models and maps to predict the likelihood and intensity of flooding has an inherent, although not immediately apparent, flaw: dependability. Flood maps and models are relatively static, updated every few years (maybe), while weather and climate are extremely dynamic. So how can static models predict dynamic events dependably? A recent study suggests they really can’t.

One of the things that make natural catastrophes inherently unpredictable is the ever-changing nature of the natural world. This is a trivial observation, but it is fundamental to understanding how peril models work and what their limitations are. This post will discuss flood specifically, but the case is the same for all natural catastrophic phenomena.