4.8
Description and use of Florida State University's high resolution historical precipitation database
Dennis D. VanCleve Jr., Florida State Univ., Tallahassee, FL; and H. E. Fuelberg and J. L. Sullivan, Jr.
A high-resolution precipitation database is useful for many purposes, including forecasting river stages, issuing flash flood warnings, and assessing pollution via streamflow models. Unfortunately, a radar only or gauge only precipitation product is not sufficiently accurate or detailed to be used confidently for these purposes. Consequently, the National Weather Service has developed the Multisensor Precipitation Estimator (MPE) software. MPE optimally combines the gauge only and radar only rainfall data onto a 4 x 4 km2 grid to create an hourly product that is more accurate than either gauge or radar only estimates. This paper will describe our use of the MPE software to generate a high resolution rainfall database for Florida, Georgia, and Alabama for the years 1996 through 2004 (9 years).
The paper will present details of the MPE process and give an overview of the rainfall climatology for the study area. In addition, we will compare results from the three different rainfall measurement schemes within the Ochlockonee River Basin of Georgia and North Florida. Besides using the MPE technique, rainfall volumes over the Basin are calculated using averages of individual gauge values as well as utilizing Thiessen polygons. The two gauge only schemes are compared with each other and with the MPE product. Rainfall volumes are calculated over the entire nine year period as well as during intervals of that total period. Differences in the rain volumes are documented using various statistical tools. Differences resulting from the following are of greatest interest: storm type (convective or stratiform), season (cool vs. warm), and the time period of the calculations (hourly, daily, etc.). Based on these differences in rain volume, we document which measurement technique is best (and how much better) under the different circumstances (e.g., convective storm, warm season, etc.).
.Session 4, Hydrologic Data Assimilation, Parameter Estimation, And Uncertainty
Thursday, 2 February 2006, 1:30 PM-5:15 PM, A403
Previous paper Next paper