modeling, rainfall and old brain

Last week my business partner, James Logan, and I spent two days in a hands-on computer workshop learning to use the U.S. Army Corps of Engineers modeling tools HEC RAS and HEC GeoRAS. It was really interesting and I think it will help both of us do a better job of understanding the challenges our engineering colleagues address routinely. We both passed, or at least they gave us each a certificate.

Our company, OneRain, supports engineering partners who build hydrologic forecast models. We provide them with the real-world data history they require to "calibrate" the river basins for which they are responsible. In particular, they need to know what sort of historical rainfall events were associated with what river responses. This history plus various model frameworks allow them to build a computational tool that can forecast what river response to expect, given rainfall information. Using HEC RAS they can forecast flows that will result downstream from various input flows, and using HEC GeoRAS they can create 3D maps that show resulting inundation patterns.

James has a B.S. in geophysical engineering from the Colorado School of Mines as part of his preparation for what we do (oh yeah, there's also that M.S. in computer science - he's well-prepared). I'm way behind that - my early background was in wetware (neuroscience; there's hardware, software, and then there's wetware, and that was my area). I was honestly worried that I would have a hard time keeping up in this seminar, having no background other than basic physics and math. I do have many years of exposure through my work to the concerns of our engineering collaborators, but having to do their work myself concerned me. There's also that general "old brain" worry - now that I've passed 50 I don't trust my brain to do things as well as it has in the past.

James pointed out that it was "interesting" to be in class with me. I would jump ahead on the problems rather than listening, and he noted afterward that more than once I asked questions which had already been answered. In the end, I had no trouble understanding the material and I found myself compelled by the tools to which we were introduced in this seminar. I now know enough to be dangerous. Nevertheless, I am very aware of what he pointed out to me, kindly but clearly; my vulnerability is to worry so much about not comprehending something correctly that I fail to listen. Lesson heard: I will do better next time.

I also learned that this isn't where the meat of what we do lies. HEC RAS and GeoRAS don't deal with rainfall at all, they're all about hydraulics and its consequences. They assume we understand everything about the hydrology already, and we're just looking at what happens to known amounts of water in the channels and the floodplain. If we want to understand how rainfall drives the hydrologic aspects of the models we're looking at, we must go further back to the data that were used to generate the hydrographs we used, in this case HEC HMS.

I'm pretty sure the methods by which the hydrographs we used in class were generated don't reflect reality, and that there isn't really a lot of disagreement about that. There also often isn't a lot of eagerness to do things differently because we're better with the devil we know than the devil we don't. As the rainfall company that's our big challenge - to come up with better ways to support these modeling approaches than what we've all been able to do to date. I believe we're just getting started on this track, although we've already made some strides with the combination of rain gages with radar rainfall data. The fact is, if we make big mistakes in understanding how much it rained to produce the hydrologic response we observe, then we are still not going to build a model that is correct for the real world.

We aren't alone in this effort, and everyone working in this domain is making steps toward improving our understanding of rainfall and how it really drives the hydrologic cycle. I am excited that my job permits me, no, requires me to continue learning and maturing my knowledge acquisition skills. I will do my darnedest to succeed so that I can achieve my goal of doing something useful. It may sound like a modest goal, but I have found it's a pretty big challenge.