We recommend not using AGCD at this point as radiation is not publicly available. ANUCLIMATE provides an open source, published alternative.
But here are a list of issues that would need to be considered:
Data is daily, will need to use a weather generator to go sub-daily (can use equations in BIOS that are provided in CABLE Fortran code. These are very basic but will do for now)
Unit conversions needed (rainfall, temperature)
Need to calculate specific humidity from 9am/3pm vapour pressure (BIOS has equations to do this)
No radiation or wind speed data available. Need to use another data source
Longwave radiation not available but can be synthesised from Swinbank equation used in CABLE (available in cable_input.F90): met%fld(:) = 0.0000094*0.0000000567*(met%tk(:)**6.0)
Precip should be taken from AGCD v2, other variables are only available in v1
We recommend not using AGCD at this point as radiation is not publicly available. ANUCLIMATE provides an open source, published alternative.
But here are a list of issues that would need to be considered:
met%fld(:) = 0.0000094*0.0000000567*(met%tk(:)**6.0)