You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've rewritten part of the IFS initial conditions, but I realized that some of the parameters are using past 12h, but I don't remember the paper mentioning that FourCastnet uses it. Are those parameters there for GraphCast and/or other models? Looking at Earth2Studio, I don't see them doing this either
channel_data= [
_get_channel(
c,
u10m=dataset_0h.u10,
v10m=dataset_0h.v10,
u100m=dataset_0h.u100,
v100m=dataset_0h.v100,
sp=dataset_0h.sp,
# ? Why is it 12h past? Should we switch it?t2m=forecast_12h.t2m,
msl=forecast_12h.msl,
tcwv=forecast_12h.tcwv,
t=dataset_0h.t,
u=dataset_0h.u,
v=dataset_0h.v,
r=dataset_0h.r,
z=dataset_0h.gh*9.81,
)
forcinchannels
]
Environment details
The text was updated successfully, but these errors were encountered:
Version
source-main
On which installation method(s) does this occur?
No response
Describe the issue
I've rewritten part of the IFS initial conditions, but I realized that some of the parameters are using past 12h, but I don't remember the paper mentioning that FourCastnet uses it. Are those parameters there for GraphCast and/or other models? Looking at Earth2Studio, I don't see them doing this either
Environment details
The text was updated successfully, but these errors were encountered: