Page tree
Skip to end of metadata
Go to start of metadata


HRES - High Resolution Forecast

The HRES is a single run created twice daily giving forecasts to Day10 based on 00 and 12UTC data times.

The HRES (current resolution 9km) provides a more detailed description of future weather than the control or individual member of the ENS.  When performance is assessed using many forecasts it provides, on average. the most accurate single-run realisation of the broadscale weather patterns.  However, any individual HRES forecast may not be the most skilful compared to ENS member forecasts, and when viewed in isolation it cannot provide an estimate of forecast uncertainty or confidence.  

It is most useful to think of HRES as a member of the ENS forecast ensemble since it contains the same physical and dynamical representation of the atmosphere and uses the same parameterisation of sub-gridscale effects.  Also, it does give a further indication of variability outside of the 51 members of the ENS by its use of higher resolution and more detailed analysis.  This brings the benefit of feeding energy from smaller to larger scale features (as happens in the real atmosphere) but with the risk of developing or amplifying spurious small features or suppressing existing ones that should be retained.  But in general on the synoptic scale, HRES and CTRL are fairly similar - the main difference is between these and the perturbed ensemble members and the relative skill of the different configurations of the IFS and their strengths has been investigated.  So far, work on combining HRES and ENS has focused on the surface weather parameters rather than general synoptic scale.  But the confidence to be placed on each depends on lead time and meteorological parameter, and can vary a lot between stations.  HRES has the advantage of higher resolution, greater detail of orography, and better discrimination of land-sea differences than ENS.  Typical HRES weighting for surface variables is such that HRES is "worth" 15-20 ENS members at Day1 and around 5 ENS members at Day10.  However, well above the surface, the HRES advantage over ENS is less significant and HRES weighting is likely to be significantly lower for forecast upper-air than for forecast surface variables since resolution and surface detail play a less significant role here

HRES is coupled to the Wave model (ECWAM). Since cycle 45r1, released in June 2018, HRES has also been coupled to the Dynamic Ocean model (NEMO). So users should be aware when inspecting model data before release of this cycle that atmosphere/ocean coupling with the HRES was not in effect.  More information on the effects of this coupling in relation to Tropical Storms is given in Appendix 12F.

Additional Sources of Information

(Note: In older material there may be references to issues that have subsequently been addressed)

  • No labels