Hi everyone,
I am currently trying to conduct a sensible heating sensitivity experiment over the Tibetan Plateau (TP) using CAM4, and I would like to ask whether anyone has done a similar experiment before.
I am mainly referring to the following two papers:
My confusion is about how this modification should actually be implemented in the CAM4 source code.
In the first paper, the authors mention modifying the vertical diffusion process by changing the parameter k. My understanding is that this corresponds to changing the diffusion coefficient in a term such as
k ∂θ/∂z.
However, if k (for example, the relevant KVT coefficient in CAM4) is multiplied by 0.5 or 1.5, I am not sure that the resulting DTV temperature tendency will necessarily become exactly 0.5 or 1.5 times the control DTV profile, because the temperature gradient and boundary-layer state may also respond to the change.
This makes me wonder how the sensitivity experiments shown in these papers were actually implemented.
More specifically:
I would greatly appreciate any advice or example code. Thank you very much for your help!
I am currently trying to conduct a sensible heating sensitivity experiment over the Tibetan Plateau (TP) using CAM4, and I would like to ask whether anyone has done a similar experiment before.
I am mainly referring to the following two papers:
- Zhao et al. (2019), Response of the Indian Ocean to the Tibetan Plateau Thermal Forcing in Late Spring
https://journals.ametsoc.org/view/journals/clim/32/20/jcli-d-18-0880.1.xml - Sun et al. (2019), Interannual Variability of the North Pacific Mixed Layer Associated with the Spring Tibetan Plateau Thermal Forcing
https://journals.ametsoc.org/view/journals/clim/32/11/jcli-d-18-0577.1.xml
My confusion is about how this modification should actually be implemented in the CAM4 source code.
In the first paper, the authors mention modifying the vertical diffusion process by changing the parameter k. My understanding is that this corresponds to changing the diffusion coefficient in a term such as
k ∂θ/∂z.
However, if k (for example, the relevant KVT coefficient in CAM4) is multiplied by 0.5 or 1.5, I am not sure that the resulting DTV temperature tendency will necessarily become exactly 0.5 or 1.5 times the control DTV profile, because the temperature gradient and boundary-layer state may also respond to the change.
This makes me wonder how the sensitivity experiments shown in these papers were actually implemented.
More specifically:
- Did they multiply the relevant KVT / vertical diffusion coefficient by 0.5 or 1.5 over the TP and then allow CAM4 to calculate DTV normally?
- Or did they first calculate DTV normally and then directly multiply the final DTV temperature tendency by 0.5 or 1.5 before adding it to the temperature tendency?
- If the goal is to obtain a vertical diffusion heating profile that is approximately 50% or 150% of the control profile, which approach would be more appropriate?
I would greatly appreciate any advice or example code. Thank you very much for your help!