Hello and thanks in advance for any help!
I am analyzing pop output from the b.e21.B1850.f09_g17.CMIP6-piControl.001 experiment contributed to CMIP6 ( as well as the 4xCO2 experiment for CMIP6 ).
I have read the discussions about the heat budget calculation on this forum as well as taking an in-depth look at the UCAR developed POP2 heat budget calculation of POP on a 0.1 degree horizontal resolution version of POP2. I'm also aware that I'm missing some output that are within that notebook.
Specifically, the POP2 output for these experiments do not have the following variables:
1. HDIFN_TEMP : horizontal diffusive flux in grid-y direction
2. HDIFE_TEMP : horizontal diffusive flux in grid-x direction
3. KPP_SRC_T : tendency from KPP non local mixing term
Using the exact code from the pop-tools workbook I have gone forward and calculated the advection terms and the solar penetration term exactly as listed in the notebook.
I have also calculated "Heat flux due to vertical mixing" in section ii) of the webpage, but I'm simply omitting the KPP_SRC_TMP variable. As the variables HDIFE_TEMP and HDIFN_TEMP are not output, iii) heat flux due to horizontal diffusion is not calculated.
My question : Is it reasonable to assume that the residual (difference in TEND_TEMP and the sum of the heat-budget terms I have) is equal to or a very good approximation for the heat flux due to horizontal mixing/diffusion as well as the KPP non local mixing?
Please see this attached figure. I have plotted time-series of the heat-budget terms averaged from 0 to 80 meters (depth of the top of the layer) in the Niño 3 region (-5°-5°N/S and 210°-270°E). Note that the "Residual" in panel (a) is calculated as the difference between the model computed temperature tendency (TEND_TEMP from POP output) and the sum of the calculated heat budget terms (Diagnostic dT/dt). All data has a 12-month running mean. Panel (b) shows the models true dT/dt and the diagnostic dT/dt without horizontal mixing/diffusion and the KPP term. Panel (c) is the same as (b), but I've added the residual from (a) to the Diagnostic dT/dt.
Thanks for any help or pointers!
Jacob
I am analyzing pop output from the b.e21.B1850.f09_g17.CMIP6-piControl.001 experiment contributed to CMIP6 ( as well as the 4xCO2 experiment for CMIP6 ).
I have read the discussions about the heat budget calculation on this forum as well as taking an in-depth look at the UCAR developed POP2 heat budget calculation of POP on a 0.1 degree horizontal resolution version of POP2. I'm also aware that I'm missing some output that are within that notebook.
Specifically, the POP2 output for these experiments do not have the following variables:
1. HDIFN_TEMP : horizontal diffusive flux in grid-y direction
2. HDIFE_TEMP : horizontal diffusive flux in grid-x direction
3. KPP_SRC_T : tendency from KPP non local mixing term
Using the exact code from the pop-tools workbook I have gone forward and calculated the advection terms and the solar penetration term exactly as listed in the notebook.
I have also calculated "Heat flux due to vertical mixing" in section ii) of the webpage, but I'm simply omitting the KPP_SRC_TMP variable. As the variables HDIFE_TEMP and HDIFN_TEMP are not output, iii) heat flux due to horizontal diffusion is not calculated.
My question : Is it reasonable to assume that the residual (difference in TEND_TEMP and the sum of the heat-budget terms I have) is equal to or a very good approximation for the heat flux due to horizontal mixing/diffusion as well as the KPP non local mixing?
Please see this attached figure. I have plotted time-series of the heat-budget terms averaged from 0 to 80 meters (depth of the top of the layer) in the Niño 3 region (-5°-5°N/S and 210°-270°E). Note that the "Residual" in panel (a) is calculated as the difference between the model computed temperature tendency (TEND_TEMP from POP output) and the sum of the calculated heat budget terms (Diagnostic dT/dt). All data has a 12-month running mean. Panel (b) shows the models true dT/dt and the diagnostic dT/dt without horizontal mixing/diffusion and the KPP term. Panel (c) is the same as (b), but I've added the residual from (a) to the Diagnostic dT/dt.
Thanks for any help or pointers!
Jacob