Work

Flow

30DayMapChallenge
Maps
Hydrology

Ceci nes pas un pipe

Day 17: Flow | #30DayMapChallenge

This is not a pipe.

When you do a lot of model QAQC, you learn to know the shortcomings of your model intimately. I was running a hydraulic model last year and was continually frustrated by how Manning’s Equation performed poorly in low slope areas. “Why are we even using such a simple model in the first place?!” I exclaimed. “Obviously this method will be lacking, it was developed to predict flow in nice clean open channels, not the mountain streams of Vermont.” Suddenly, “This is not a pipe” popped into my head, a reference to Rene Magritte’s 1929 painting The Treachery of Images. The wordplay gave me a little chuckle, but it also struck a deeper chord. My model of a river was no closer to being a river than Magritte’s image of a pipe was to being a pipe.

When we see the shortcomings and rough edges in the output of a simpler model, we are reminded to take all of its predictions with a grain of salt. As a young engineer, I saw the impressive results from a 2D HEC-RAS model and was lulled into a somewhat false sense of security. Seeing is believing, it seems. These days, I’m seeing models more as places to build intuition and investigate what could be, rather than mirrors of reality. Whenever the topic of the limits of modeling comes up now or a peer recites “all models are wrong, some are useful”, I’m reminded of Magritte and his pipe.

30DayMapChallenge: https://30daymapchallenge.com/