In earlier post I mentioned how very different the ECMWF and GFS forecasts were for 00 UTC on 21 November. Here is a look at the final outcome. The forecast from ECMWF for 500 mb is above and the same for the GFS is at bottom. In the middle is the NAM 500 mb analysis for 00 UTC on the 21st - last evening.
Very hard to make a call as to which model was better. The ECMWF was clearly closer to reality over eastern North America. Western North America and eastern Pacific were handled badly by both models. The ECMWF was seriously too intense and too far north with the Gulf of Alaska system, while the GFS was about the right intensity, but was too far south and west. The GFS was much in error over Alaska; whereas the split in the flow over Northwest and bowling-ball short-wave for Southwest were purely fiction in the ECMWF.
So, just a good illustration of the inherent dangers of making routine seven-day forecasts!
No comments:
Post a Comment