Upgrade WRF and WPS for new GFS

I got the following in an email from WRF:

Beginning 1200 UTC July 19, NCEP will upgrade GFS, as well as several output datasets provided on its ftp site. The updates, which include adding missing values to land fields (such as soil temperature, moisture and snow etc.), and changing the landsea mask field, will break the WRF model if you are not using an updated version of WPS.

…which I ignored. Then my WRF forecasts stopped working, so I upgraded to the new WRF/WPS.


Install CDO

CDO is a great piece of software for working with GRIB and NetCDF data:

  1. Install Libs4cdo:
  2. Install cdo



Journal Club: 30-05-2017

Article title: Two-Dimensional Indices of Atmospheric Blocking and their statistical relationship with Winter Climate Patterns in the Euro-Atlantic Region.

(Published in the International Journal of Climatology in 2006)

This article identified some statistical linkage existing between climate variability patterns and the frequency of atmospheric blocking over the Euro-Atlantic region.

The research of this article started by defining three different blocking indices, i.e., indices that capture the occurrence of blocking episodes according to a given set of atmospheric conditions over time. One of these blocking indices is based on meridional gradients of Geopotential Height. The other two are based on vertical gradients of Potential Vorticity. The different characteristics of these (and other) blocking indices offer more robustness to the results shown by the research.

Those indices, applied to a gridded dataset, can express the frequency of blocking over the respective geographic domain. These types of outputs can be compared with other spatial features, in an attempt to find any links that help describe blocking episodes.  This paper assessed blocking frequency against the indices of the Winter climate patterns of the region, following tips by Pavan et al (2000) and Stein et al (2000); and better described not only the relationship between blocking and the North Atlantic Oscillation, henceforth NAO (primary mode of variability in the region), but for also climate patterns. In sum:

  • Anti-correlation between the NAO index signal and blocking frequency in the Atlantic.
  • Correlation between the NAO index signal and blocking frequency in Central and Western Europe.

This shows that the NAO-blocking relationship strongly depends on location.

Blocking frequency and the NAO: positive phase of the NAO on the left (negative on the right). Dark (light) grey denotes areas where there is significantly more (less) blocking according to the NAO phase in analysis.

Blocking occurrence was also associated with other Winter climate variability pattern, specially, the secondary (EOF2) and tertiary (EOF3) climate patterns (sometimes called the East Atlantic pattern and the Scandinavian pattern). The authors highlight how the Scandinavian pattern (EOF3) is consistently related with blocking over the North Sea and the Norwegian sea.


Relevancy for our research:

– If atmospheric blocking in this region is, to some extent, dependent on the various Winter climate patterns’s phase, then some success in forecasting atmospheric blocking may be attained by further understanding the Winter climate patterns, some of which have been shown to be somewhat predictable months ahead.

– Understanding if, and how, peaks/lows of renewable energy generation (e.g. wind and solar) are related to blocking events may help improve energy forecasting.



Pavan et al. (2000)– “Seasonal prediction of blocking frequency: Results from winter ensemble experiments”

Stein (2000) – “The variability of Atlantic–European blocking as derived from long SLP time series”




Wind Energy Science Conference 2017

I had a great trip to the Wind Energy Science Conference 2017 last week (28 June 2017) in DTU, Copenhagen, where I saw lots of interesting presentations about wind power forecasting.

I gave a talk about our ESIPP research, comparing wind and solar observation data to reanalysis data and forecast data. I made the point that reanalysis data are not always in good agreement with local observations, and that sometimes forecasts at a higher resolution can outperform reanalysis at a lower resolution, even though the latter are driven by observations. This shows the importance of surface-driven (high-resolution) data on skill.


Presentation slides are available here: WESC2017 presentation