Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

Cold Points at coastal areas.


New member

I successfully ran the WRF with data from the test ECMWF. I noticed that when I plotted the temperature at 2 m, there are cold spots on the coasts.

When I compared the Land-sea mask used by the geogrid and the one used by the ECMWF (Figure A, gray) I noticed that they are not compatible and the white dots show the location of the cold spots that you can see in Figure B. I was able to minimize them by changing the interpolation in METGRID.TBL, w_16 point average being the one that has worked best but not 100%.

Do you know of any tween that can help me minimize or fix the problem? or Any other ideas that could help me?




  • Figure A.JPG
    Figure A.JPG
    18.9 KB · Views: 3
  • Figure B.JPG
    Figure B.JPG
    32.5 KB · Views: 3
I´m using this data:
Test data for the US Weather Research and Forecasting (WRF) model - Data and Charts - ECMWF Confluence Wiki
These are data tested by the ECMWF who have worked on the WRF. Currently FTP is no longer available, but the data I downloaded covers much of Central and North America.

My first hypothesis is that the Land-mask of the geogrid (It comes from the UCAR data Static Data Downloads) is different from the one handled by the ECMWF; then these differences are the ones that metgrid, when interpolating, considers them as water data and puts colder points on the coast.