The key to being prepared is time. As reported by The Weather Channel, Deyn Johnson and her husband evacuated their Boulder Creek home only when their cat, Jezebel, jumped on her sleeping husband at 4:30 in the morning, “batting at him and yowling.”
Scientists prefer, through models, to be able to provide forecasts with significantly longer lead times than that, but flash floods in small, steep, drainage basins, by their nature don’t always allow it. Predicting riverine flooding (flooding in a floodplain area), with predicted crests, is an easier task.
At the time of writing this report, the USGS’s ‘Flood Inundation Mapper’ was showing the continuation of major flash flooding in Colorado, and riverine flooding in North Dakota.
Types of Floods
The Susquehanna River Basin Commission has stated that “the only thing two floods have in common is water.” In reality, however, we do know a lot of general facts about floods including the types:
- Dam failure: When a dam fails, the down-stream flooding is swift and can pull a significant amount of debris into the water.
- Storm surge or coastal flooding: When there are intense low-pressure systems over the open ocean, water levels elevate above the normal tidal range.
- Riverine flooding- high water levels over top of the natural- flood plain- or artificial banks of a stream or river. Note: a predictable fixed time ahead; and
- Flash flooding as in Colorado- when there’s too much rain for the soil to absorb – is usually caused by slow-moving thunderstorms. Note: development in six hours or less from rainfall to the onset of flooding; and are the number one weather-related killer in the U.S.
Floods and the USGS Mapper
Generally, the degree of natural water inundation is related to creek and river levels which in turn are measured using stream gauges. The recorded data is then stored onsite and later transmitted to USGS offices for daily comparison mapping and graphing.
The stream flow data is then computed into percentiles based on the period of record for the current day of the year. For instance, as the USGS explains ”a river discharge at the 90th percentile is equal to or greater than 90 percent of the discharge values recorded on this day of the year during all years that measurements have been made.”
When a river rises and overflows its banks, the water spreads over the flood-plain. In the case of most flooding or inundation, a percent exceedance is used, calculated by subtracting the percentile scale value from 100 (100-75=25).
The result of all of this is a considerable amount of data collected on a daily basis for the purpose of flood forecasting.
In the case of Boulder Creek, however, as noted by NOAA, forecasts for the Boulder Creek near Orodell are issued as needed during times of high water, but accurate data is not routinely available. One reason for this is that the gauge readings are, and were affected by the upstream reservoir operations.