jsbj@uchicago_edu
New Member
Hi CESM forum,
I attempted to run CAM5.3 at T31 resolution with a slab ocean, CAM5 physics, and prescribed bulk aerosols to equilibrium under 1, 2, and 4 times preindustrial CO2 values. Under 4xCO2, the model crashes in the 14th year, likely due to a CFL violation. I reran the model with a smaller time step (dtime = 600). In order to test whether changing the time step affects the climate directly, I reran the 1 and 2xCO2 cases with this smaller timestep, and something strange happened – the model equilibrated when the top-of-model energy balance (FSNT-FLNT) was ~1 W/m2 as opposed to 0 W/m2, and at a surface temperature 2ºC lower than when dtime = 1800 (see the Gregory plots here).
I was wondering if you folks had seen anything like this before, and had any thoughts as to why I might be getting such a strange outcome (is it even possible for the model to equilibrate with FSNT-FLNT = ~1 W/m2?). In the meantime I'll investigate specifically how the model behaves differently with the shorter time step.Many thanks!
Jonah
I attempted to run CAM5.3 at T31 resolution with a slab ocean, CAM5 physics, and prescribed bulk aerosols to equilibrium under 1, 2, and 4 times preindustrial CO2 values. Under 4xCO2, the model crashes in the 14th year, likely due to a CFL violation. I reran the model with a smaller time step (dtime = 600). In order to test whether changing the time step affects the climate directly, I reran the 1 and 2xCO2 cases with this smaller timestep, and something strange happened – the model equilibrated when the top-of-model energy balance (FSNT-FLNT) was ~1 W/m2 as opposed to 0 W/m2, and at a surface temperature 2ºC lower than when dtime = 1800 (see the Gregory plots here).
I was wondering if you folks had seen anything like this before, and had any thoughts as to why I might be getting such a strange outcome (is it even possible for the model to equilibrate with FSNT-FLNT = ~1 W/m2?). In the meantime I'll investigate specifically how the model behaves differently with the shorter time step.Many thanks!
Jonah