A new verification approach for nowcasting based on intensity and spatial-temporal feature correction.

Area of Science:

• Meteorology
• Atmospheric Science
• Weather Forecasting

Background:

• Conventional nowcasting verification methods for strong convective weather suffer from double punishment and low scores.
• Existing methods lack the ability to fully capture the nuances of convective events, limiting forecast potential.

Purpose of the Study:

• To propose an improved neighborhood verification method for strong convective weather nowcasting.
• To address the limitations of traditional "point-to-point" verification by incorporating spatial, temporal, and intensity information.
• To enhance the evaluation of nowcasting accuracy and usefulness.

Main Methods:

• Developed a neighborhood verification method based on the spatial neighborhood fraction skill score (FSS).
• Designed intensity and time neighborhood correction schemes, including a comprehensive approach.
• Introduced spatial neighborhood probability and fuzzy logic with a Gaussian membership function.
• Implemented an adaptive weighting strategy for time neighborhoods to handle forecast timing errors.

Main Results:

• The proposed neighborhood verification method increases verification tolerance compared to traditional indicators.
• It provides more informative evaluations by considering spatial scale, time, and intensity.
• The method demonstrates better alignment with the practical needs of strong convective weather nowcasting operations.

Conclusions:

• The novel neighborhood verification method offers significant advantages over conventional techniques for strong convective weather nowcasting.
• It provides a more comprehensive and realistic assessment of forecast performance.
• This approach enhances the technical development and operational utility of nowcasting systems.