17th International Conference on Interactive Information and Processing Systems (IIPS) for Meteorology, Oceanography, and Hydrology

1.10

Developing local Quantitative Precipitation Forecast (QPF) using neural networks

Keith M. Stellman, NOAA/NWS, Slidell, LA; and J. Kuhn and J. Graschel

Great emphasis has been placed on quantitative precipitation forecast (QPF) to improve lead time in river forecasting. The hydrologic model at the Lower Mississippi River Forecast Center (LMRFC) uses 24 hour QPF twice a day at 12 UTC and 00 UTC. The location and timing of QPF can have a major impact on river forecasting. In order to improve forecasting QPF, a neural network was developed using a software program called Brainmaker.

Upper air data from local soundings and precipitation are used to train the neural network. Local programs are used to decode and calculate parameters from the sounding and are ingested into Brainmaker. Hourly and daily precipitation are automatically collected and used in the neural network. The presentation will show our results from the neural network on QPF forecasting.

Session 1, IIPS advancements/applications in Forecasting and Observation System Technologies, Climatology, Oceanography, and Hydrology (Parallel with Session 2, 3, J1, & J2)
Monday, 15 January 2001, 8:30 AM-5:15 PM

Previous paper  Next paper

Browse or search entire meeting

AMS Home Page