Since this seems to be a week of heck and high water, I thought I’d return to the topic of floods. We’re right at that time of year when we start to see flooding from two different kinds of events, one more predictable than the other. There are flash-floods caused by too much precipitation, snow-melt floods, and I suspect urban flooding caused by impermeable surfaces. Snow-melt floods used to be called “the spring rise” and are relatively predictable.
What makes a flood? When more water comes than the soil can absorb. This may be due to paving or otherwise hardened ground that can’t absorb the rain, due to snow melt that saturates the ground and so starts to run overland and fill streams, or rain that falls so quickly that the ground just can’t absorb it fast enough and it “runs off” the surface and into streams, or puddles and spreads and puddles and spreads and your street is now a canoe route. Sometimes you get all three at once. I call that “upper Midwest between March and June.”
Of the three, snowmelt flooding is the easiest to predict, especially once you develop a data set that covers several decades of snowfall, date of thaw starting, and river gauges. If you don’t have much snow, then there’s not going to be much of a rise, even though it might melt quickly. What makes life interesting is rain on top of lots of snow while the ground is still frozen, then a rapid thaw. The frozen soil can’t absorb much of the water, the rain melts the snow, then the fast thaw really melts the snow. It flows very quickly into streams and rivers, swelling them and causing headaches the farther downstream you go. What hit the valley of the Red River of the North in the late 1990s was a worst-case scenario of heavy winter snow, rapid thaw upstream, and rain. Grand Forks flooded and caught fire, and then Winnipeg and other downstream cities got hit. Instead of the downstream land thawing and being able to absorb the water from upstream more easily, since the Red River of the North flows northwards, the thaw moved with the floodwaters. Hydrologists are still writing papers about it.
It’s a bit harder to predict single-storm flash flooding, although technology has improved the odds. When you get ten inches of rain in an hour, it is going to flood. That’s just a given, unless it falls on pure sand, and even then you’re going to see stream overflows right quick pronto. If the ground is already saturated from earlier rain, it will take less moisture to cause problems, and you can predict that “if we get much more, the creek’s gonna rise.” The storm that triggered the Playa That Ate A Neighborhood left three and a half inches of rain in an hour, and the ground was moist but not saturated. Hello, water in front yards and streets!
Rain on hard surfaces, the main source for desert floods or “urban flood advisories” is pretty self-explanatory. Rain in a rock is going to run off, unless said rock is pumice or another very permeable type. Rain on cement, asphalt, shingles, is going to run off and not be absorbed. This is why developers get surprised when places that never flooded before suddenly have street flooding. You cover the soil with fake rock, and guess what happens? If you know anything about water, the events are predictable. Amarillo is very predictable, to the point that when a storm comes, people automatically began planning routes that avoid about five intersections in town. If we get more than two inches, four more intersections get added to the “do not attempt” list. Seriously, if the high-water gauge is showing six feet of water, even your jacked-up super-off-road pick-up is not going to make it. Trust me on this.
Right now, we’re seeing flooding due to run-off due to heavy rain on baked soil (CA), excess rainfall beyond what the soil can hold (CA, the southeast, OH), and by the end of the month the first snow-melt river rises will begin.
“Rivers flood. That’s what rivers do,” as one retired farmer told me. He’d lived on the river most of his life, and just took it for granted that floods happened.