I am looking at the secant bulk modulus calculation in the nonlinear Equation of State. As pointed out in the comments of the code, it is done following the polynomial fitting of Jackett and McDougall (1995). However, the coefficients used in the code (ROMS 2.1) are very different from that appeared in that paper. Could someone explain why there is such difference?
Thanks,
Haosheng Huang
UMassD
secant bulk modulus calculation in the nonlinear EOS
-
arango
- Site Admin
- Posts: 1351
- Joined: Wed Feb 26, 2003 4:41 pm
- Location: DMCS, Rutgers University
- Contact:
In ROMS/TOMS, it is assumed that there is not pressure variation along geopotentials, so pressure and depth are interchangeable 
The coefficients associated with pressure are divided by 10 and the ones associated with squared-pressure are divided by 100. Recall that to convert from bar to meter we use:
pressure (bars) = 0.1 * depth (meters)
Also notice that when comparing the model equation of state coeffiecients with those provided by Jackett and McDougall (1995), the sign of some of the coefficients have been changed so the expasion contribution for each term is always added. This facilitates, for example, taking in tangent linear and adjoint transfomation for these equations. We still get the same density related quatities. Notice that rho_eos.F includes several check values for this equation of state. This equation of state is only valid for a potential temperature range of -2C to 40C and salinity range of 0 PSU to 42 PSU. This is an equation of state for potential temperature and not in situ temperature. However, the model density rho used in the pressure gradient routines is in situ density. See Jackett and McDougall (1995) for more details. The temperature in ROMS/TOMS is always potential temperature.
The coefficients associated with pressure are divided by 10 and the ones associated with squared-pressure are divided by 100. Recall that to convert from bar to meter we use:pressure (bars) = 0.1 * depth (meters)
Also notice that when comparing the model equation of state coeffiecients with those provided by Jackett and McDougall (1995), the sign of some of the coefficients have been changed so the expasion contribution for each term is always added. This facilitates, for example, taking in tangent linear and adjoint transfomation for these equations. We still get the same density related quatities. Notice that rho_eos.F includes several check values for this equation of state. This equation of state is only valid for a potential temperature range of -2C to 40C and salinity range of 0 PSU to 42 PSU. This is an equation of state for potential temperature and not in situ temperature. However, the model density rho used in the pressure gradient routines is in situ density. See Jackett and McDougall (1995) for more details. The temperature in ROMS/TOMS is always potential temperature.
-
huanghaoshen
Dear Hernan,
Thank you very much for your reply.
I am fully aware of the points you mentioned in your reply. However, I am still comfused with the parameter values used in the code. For example, the constant used in secant bulk modulus in Jackettand McDougall (1995) is 1.965933e+04 while the same constant used in ROMS (in mod_eoscoef.F) is A00=1.909256e+04_r8. Do I misunderstand something in comparing these two values?
Haosheng
Thank you very much for your reply.
I am fully aware of the points you mentioned in your reply. However, I am still comfused with the parameter values used in the code. For example, the constant used in secant bulk modulus in Jackettand McDougall (1995) is 1.965933e+04 while the same constant used in ROMS (in mod_eoscoef.F) is A00=1.909256e+04_r8. Do I misunderstand something in comparing these two values?
Haosheng