Flash Flood Guidance issued by the National Weather Service-Past, Present, Future

Author(s): Costanza, K.

Flash flooding is a serious threat that accounts for the largest number of weather related deaths per year in the United States. The National Weather Service realized the severity of this threat during the Independence Day flooding event of 1969 which killed 41 people over a few counties in Ohio. That event sparked the National Weather Service to begin issuing some form of flash flood guidance that could effectively warn communities of potential flash flooding risks associated with a rain event. Over the years, the models for determining the flash flood guidance values have evolved from simple "rules of thumb" to a more scientific basis.

The current model, the Gridded Flash Flood Model (GFFG), used to determine flash flood guidance is based on a 4km by 4km grid scale and uses the National Resources Conservation Service (NRCS) Curve Number methodology. This method was chosen because of its ability to take into account the antecedent soil moisture conditions of a system, calculate the abstractive losses based on a Curve number, and calculate a peak flow by way of the Triangular Unit Hydrograph method. The determination of the antecedent soil moisture conditions are determined by a distributed hydrological model and relayed to the GFFG Model. The NRCS Curve Number method is also appealing because it can be tied to the physical world through the determination of a curve number which can take into account the spatial variability of soils types, vegetative cover and slope of a watershed. The final guidance of the model is varying rainfall amounts associated with the appropriate temporal scales (1, 3, 6, 12 and 24 hour) likely to cause flooding for an area.

Due to the serious nature of flash flooding, the need for improvement to the current model is imperative. Although the current model is a drastic improvement relative to the past models, there is still room for further improvement. One example would be incorporating finer spatial resolution data, such as soil data, to determine new Curve Numbers used in the model. With current GIS applications, the incorporation of this type of data is relatively simple. Other more advanced improvements could include evaluating different infiltration models to determine the abstractive losses of the system. The methods used in the current model are strictly empirical and the use of more physically based infiltration models could produce better results. In addition, the current model lacks connectivity between grid cells which could cause issue if there is a rain event upstream of a "problem cell". This connectivity could be gained by routing flow from one cell to another. Continued scrutiny of the current model will only yield improved guidance issued to communities resulting in more credibility of products rendered by the National Weather Service.

Go back


Technical Sessions

Session I Sedimentation
Session II Weather/Climate
Session III Coastal Resources
Session IV Surface Water Management
Session V Wetlands
Session VI Education
Session VII Management/Planning
Session VIII Wetlands
Session IX Delta Groundwater
Session X Nutrients
Session XI Delta Water Resources
Session XII Ports

Click here to view the entire agenda