Session 4.8 Modeling the Tallahassee Minimum Temperature Anomaly

Tuesday, 21 June 2005: 4:30 PM
South Ballroom (Hilton DeSoto)
Kelly G. Godsey, NOAA/NWS, Tallahassee, FL; and H. E. Fuelberg, J. J. O'Brien, A. I. Watson, and R. L. Block

Presentation PDF (1.8 MB)

During the last three years, a joint study on minimum temperatures has been conducted by the National Weather Service (NWS) in Tallahassee and The Florida State University Department of Meteorology. This project has investigated the meteorology of the Tallahassee, Florida area by establishing a network of observers throughout the urban and rural area. The observers record morning minimum temperatures and submit them monthly to a data coordinator who calculates basic statistical quantities and prepares reports describing results of the temperature study.

The overall goal of the study has been to provide operational forecasters at NWS Tallahassee with a model providing site specific daily minimum temperature guidance. One important use of this guidance will be to assist with assessing the need for freeze warnings. To that end, the three years of data have been analyzed using an empirical orthogonal function (EOF) method. The EOF analysis is a statistical procedure designed to extract the dominant spatial patterns that are present within any data set. The EOF technique has provided insight into the mechanisms driving the Tallahassee temperature anomaly. It also has allowed the local urban heat island to be mapped and a minimum temperature forecast equation to be developed for operational forecasting.

This forecast model, rooted in the three year climatological database for each site, is found to perform best during ideal radiational cooling conditions. The model has limited skill at predicting minimum temperatures under less than optimal conditions.

This paper will describe our data set, the local urban heat island that is detected, the minimum temperature guidance equations that are developed, as well as an analysis of their results. The technique described is relatively simple and could be applied to many other locations around the country.

In the months ahead, this model will aid operational forecasters at NWS Tallahassee in predicting dangerous freezes that can cause significant problems for residents and agricultural interests in Tallahassee and the immediate vicinity.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner