Laserfiche WebLink
<br />OOO~84 <br /> <br />4 <br /> <br />sica 1 processes as soc i ated with flood event s. Input data to the model are <br />retrieved from a separate data management component of the system and results <br />of the simulation modeling are then returned to this data manager. <br />One of the ongoing maintenance tasks associated with simulation models is <br />calibration and verification, Performance of the simulation models should be <br />continually assessed and adjusted to be consistent with measured rainfall and <br />runoff. If inconsistencies develop from calibration to calibration (i.e. one <br />event under estimated followed by an over estimate) the model should be <br />reviewed to see if all important physical processes are in fact being simu- <br />lated or if a simplified routine needs to be enhanced. Review of model per- <br />formance should take place after all major rainfall/runoff events. <br /> <br />Data Storage and Management <br />The capability for data storage and management is central to the opera- <br />tion of the flood-warning system, Because of the variety of data that is <br />required to provide a timely forecast and also the speed at which new data <br />must be incorporated into the system, an efficient method for handling data is <br />important. Conceptually, the data management system can be thought of as the <br />"motherboard" into which all of the other components of the system plug. The <br />data management systems provides access to the data base for the flood-warning <br />system that contains all the historic and real-time data needed to perform the <br />forecasting and warning tasks, <br />The data acqui sit i on system feeds data to the data management system <br />which in turn stores the data in a predefined format, Simulation models <br />retrieve data using the data management system and then return results to the <br />data base, Utility programs access the data management system to perform such <br />tasks as preparing an updated forecast or displaying current data. While it <br />may seem cumbersome to have all data carried out via a data management system, <br />access to the data base is, nevertheless, very fast. Use of a data management <br />system assures the integrity of the data base and protects the data base from <br />unintentional or intentional tampering. Figure 1 shows an example of a <br />centralized data management system as implemented by the Corps of Engineers <br />Hydrologic Engineering Center for use with their real-time water control soft- <br />ware. (7) <br />