Determining an Optimal Decay Factor for Bias-Correcting MOS Temperature and Dewpoint Forecasts

The National Centers for Environmental Prediction (NCEP) has for several years bias-corrected upper air forecasts by using a simple technique called “decaying average” (Cui et al. 2012). MOS forecasts are unbiased over the sample period of development. However, biases within that period can exist, and application to new data may produce forecasts with short or long term biases. Because NWP models undergo change over time, MOS forecasts, unless redeveloped, can be biased because of these changes, and also possibly because of local environmental changes. In the decaying average method, a “decay factor”alpha is used. The value of alpha affects not only the short and long term bias characteristics, but also other accuracy measures of the forecasts. This paper will show how different values of alpha affect MOS temperature and dewpoint forecasts, and the range of alphas that would be appropriate for bias correcting those forecasts. Biases and other quality measures will be shown for both cool and warm season samples before and after various values of alpha have been applied. The paper will also compare the decaying average method to a regression method used in the AWIPS application Boise Verify, and will discuss issues in possible implementation of the two methods.

Reference:

Cui, B., Z. Toth, Y. Zhu, and D. Hou, 2012: Bias correction for global ensemble forecast. Wea. Forecasting, 27, 396-410.