2014•Unpublished venueRequires access

Sensitivity of 96 and 120-hour Numerical Model Tropical Cyclone Position Forecasts to Initial Position Errors

Coy C Fischer

Open publisher page 0 citations

Abstract

Abstract : Global Forecast System (GFS) western Pacific tropical cyclone track forecasts from the 2011, 2012, and 2013 seasons (87 storms) were compared to Joint Typhoon Warning Center (JTWC) tropical cyclone best track data and warning bulletins in order to determine the sensitivity of 96 and 120-hour tropical cyclone position forecasts to initial position error. A tropical cyclone vortex tracker, which uses seven different model parameters to track storm centers, was implemented to determine model forecast positions. The differences between JTWC analysis positions and the model-derived vortex positions were analyzed at each forecast hour (00, 12, 24, 36, 48, 72, 96, and 120). The relationship between the geographical spread among the model vortex-tracking parameters and forecast errors was also considered. Correlations between error at each forecast hour and the initial 00-hour error suggest that position error has no effect on forecast error at 96 and 120-hours.

About this research paper

What this paper is about

Abstract : Global Forecast System (GFS) western Pacific tropical cyclone track forecasts from the 2011, 2012, and 2013 seasons (87 storms) were compared to Joint Typhoon Warning Center (JTWC) tropical cyclone best track data and warning bulletins in order to determine the sensitivity of 96 and 120-hour tropical cyclone position forecasts to initial position error. A tropical cyclone vortex tracker, which uses seven different model parameters to track storm centers, was implemented to determine model forecast positions. The differences between JTWC analysis positions and the model-derived vortex positions were analyzed at each forecast hour (00, 12, 24, 36, 48, 72, 96, and 120). The relationship between the geographical spread among the model vortex-tracking parameters and forecast errors was also considered. Correlations between error at each forecast hour and the initial 00-hour error suggest that position error has no effect on forecast error at 96 and 120-hours.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Abstract : Global Forecast System (GFS) western Pacific tropical cyclone track forecasts from the 2011, 2012, and 2013 seasons (87 storms) were compared to Joint Typhoon Warning Center (JTWC) tropical cyclone best track data and warning bulletins in order to determine the sensitivity of 96 and 120-hour tropical cyclone position forecasts to initial position error. A tropical cyclone vortex tracker, which uses seven different model parameters to track storm centers, was implemented to determine model forecast positions. The differences between JTWC analysis positions and the model-derived vortex positions were analyzed at each forecast hour (00, 12, 24, 36, 48, 72, 96, and 120). The relationship between the geographical spread among the model vortex-tracking parameters and forecast errors was also considered. Correlations between error at each forecast hour and the initial 00-hour error suggest that position error has no effect on forecast error at 96 and 120-hours.

Key concepts: Tropical cyclone, Typhoon, Tropical cyclone forecast model, Meteorology, Track (disk drive), Climatology, Position (finance), Environmental science

Related papers

Back to paper searchBrowse research topicsOriginal source
Sensitivity of 96 and 120-hour Numerical Model Tropical Cyclone Position Forecasts to Initial Position Errors — Research Paper | ScholarLens