In the scenarios paper in 2020, we switched to scaling seasonality change with NPP from MAGICC rather than this composite temperature-CO2 timeseries that was used for historical. I wonder if we should switch to being consistent throughout. We don't really have NPP back in time, which is an issue, but maybe less than it first appears because seasonality change is kept constant pre-1850 anyway.
In the scenarios paper in 2020, we switched to scaling seasonality change with NPP from MAGICC rather than this composite temperature-CO2 timeseries that was used for historical. I wonder if we should switch to being consistent throughout. We don't really have NPP back in time, which is an issue, but maybe less than it first appears because seasonality change is kept constant pre-1850 anyway.