A flood warning system (FWS) can save lives and reduce property damage by increasing mitigation time--the time available to take proactive actions prior to the occurrence of high water. To achieve this, a FWS measures rainfall and streamflow. By examining these measurements, an analyst can identify hazardous situations and notify the public or public officials, giving them time to respond. Intuitively, more data measured with more gages seems desirable for such a system. But gage installation and maintenance is costly. Furthermore, the information that can be obtained from a gage network is limited: each additional gage does not contribute equally to increasing the mitigation time. Thus an economic trade-off exists. One critical task in FWS planning is to consider this trade-off to select the optimal number, type, and location of gages for a system.
The National Weather Service (NWS) asked us to develop a procedure to estimate, for a flood warning feasibility study, the economic value of a gage network in a watershed. The procedure developed should provide reproducible results and be applicable for a wide range of watersheds. Factors such as watershed characteristics and gage type and density should be included in the benefit estimator.
|
We developed a procedure that: (1) relates the number and type of gages to error in precipitation estimates and to error in evaluation of the flood hazard; (2) relates these errors to their impact on mitigation time; (3) relates mitigation time to economic benefit.
We developed the benefit evaluation procedure through numerical experiments with a watershed forecasting model in a Monte Carlo sampling paradigm. If combined with a cost estimate for a gage network, the results of application of the procedure will provide information for the required trade-off analysis.
The results will be used by NWS staff to evaluate proposed flood warning systems, thus permitting a more rigorous evaluation of the benefits achievable. This, in turn, will provide information for wiser investment.
|