How was the Graphcast Small 1.0 degree Reanalysis made?
tvonich opened this issue · 1 comments
tvonich commented
Hello,
Was it simply e quarter degree reanalysis values pulled from whole number lat/lon positions or were the values actually averaged over the 1.0 degree region?
alvarosg commented
Thanks for your message.
It was done by bilinear interpolation, which for this grid is indeed equivalent to pulling the values at the whole lat/lon positions.
It was done this way so the 1 deg values would match the statistics of the 0.25 deg values, and the metrics could be compared across resolutions.
Hope this helps!