In my last blog post, I wrote briefly about the 100-year storm. It’s a nuanced term with everyday use in water management circles, but is often one which causes confusion when presented in public view, deserving of some unpacking here.
In simple terms, this nomenclature – describing storms as a 2-year, 5-year or 100-year storm and so on – is shorthand for describing a storm’s rarity and relative size in the context of historical climate data. A 100-year storm is a rainfall event which has a 1% probability of occurring every year. A 2-year storm is one which happens much more frequently.
Probabilistic, not predictive
The confusion arises when some misconceive a 100-year storm will only happen once in 100 years. In actuality, 100-year storms can happen at any time – in back-to-back years or back-to-back days. It would be inaccurate to think, for example, since Austin experienced a 100-year storm in 2010 – as it did – the next 100-year storm would only happen in 2110.
To illustrate this further, let’s shift paradigms. In the world of Texas Hold ‘Em Poker, a player is initially dealt two cards called “pocket” cards. When dealing from a freshly shuffled deck, there are 220-to-1 odds you will be dealt two “pocket” aces – a very rare deal. Collecting the cards and repeating the process yields the same odds of receiving the same deal. The probability of being dealt two aces in back-to-back hands is, indeed, quite unlikely, but it can happen. In the same way, two large, statistically rare storms can occur in succession, but that succession would also be quite unlikely.
More than one 100-year storm
Keep in mind storms are defined by two qualities: the duration of rainfall and the depth of rainfall. As shown in the table below, Austin could experience a 100-year storm with any combination of these parameters. Storm A, which lasted 30 minutes and deposited 3 inches of rain has the same probability of occurring as Storm B, which lasted 24 hours and dropped 10 inches of rain. We know this from our statistical analysis of climate data. Both Storm A and Storm B are 100-year storms but the water elevations in Shoal Creek in response to these storms would be quite different.
City of Austin
|2-year||1.32 in||1.72 in||2.67 in||3.44 in|
|5-year||1.71 in||2.28 in||3.56 in||4.99 in|
|10-year||1.98 in||2.68 in||4.21 in||6.10 in|
|100-year||3.04 in||4.37 in||6.85 in||10.20 in|
1. Values obtained from City of Austin Drainage Criteria Manual
2. Climate data used to derive these values obtained from climate station at Tom Miller Dam
Designing for the 100-year storm
The confusion gets worse when water professionals get lazy in speech. When we refer to the 100-year storm, we mean the 100-year design storm – the storm used to model watersheds, set regulations and design infrastructure. Many cities in North America require underground infrastructure such as storm sewers be able to convey a 5-year storm flow and use a 100-year storm when modeling surface features like watercourses or flood setbacks. The City of Austin uses the 24-hour 100-year storm as their standard design storm. As noted in my last blog, the City’s Watershed Protection Department estimates ~275 structures are at risk of inundation during the 100-year storm when Shoal Creek is modeled with the 100-year design storm.
- The “100-year” nomenclature defines a storm based on its probability of occurrence which we know from examining historical climate data
- A 100-year storm is simply a rainfall event which has a relatively low probability of occurring
- The 100-year design storm for the City of Austin is a 24-hour event which deposits 10 inches of rain
If you have any watershed management topics you’d like me to unpack in this blog, post your question in the comments below!