The purpose of this project is to verify the official forecasts and the different models used as guidance by the NWS in Miami, FL, after the implementation of the Southern Region Grid Policy in November 2014. It is believed that by using forecasts blends the official forecast can be made with an improved accuracy. The data is obtained from the AWIPS II system which houses numerous models for use with BOIVerify, which Jenks defines as a forecast weighting system that incorporates bias-corrected versions of the direct and statistical model output, and then compares the mean average error (MAE) of each of them. In our analysis, we compared the maximum and minimum temperature for the winter season of 2014-2015. Our results show that the National Weather Service in Miami, FL is already generally matching or outperforming the SuperBlend model in both the maximum and minimum temperature fields. In addition we compared model probability of precipitation forecasts (PoP) with the NWS’s PoP forecasts and found that the NWS overestimates low PoP values and underestimates high PoP values. Hopefully this study can be used to improve forecasts of high temperature, low temperature, and precipitation for the South Florida Forecast Region.
Byrd, Jacob, "The use of BOIVerify at the National Weather Service in Miami, FL to enhance the accuracy of temperature and precipitation forecasts" (2015). Internship Reports (Restricted). 109.
For UM Patrons Only