Refactors gas exchange carbon chemistry (+ adds some complexity) #186
Conversation
jagoosw
changed the title
|
@bethcandish you could do some testing with the "best practice" parameters with this branch and see if it makes any difference, and if so change them here. |
…ed gas exchange docstring
| @@ -151,12 +151,12 @@ resulting in the equilibrium constant: | |||
| K_B = \frac{\ce{[B(OH)^-_4][H^+]}}{\ce{[B(OH)_3]}}. | |||
| ``` | |||
|
|
|||
| Finally, taking the DIC and Alkalinity in micro equivalents (i.e. scaled by ``10^{-3}/\rho_o`` from mmol C/m``^3``) denoted by ``\bar{DIC}`` and ``\bar{Alk}``, the carbonate alkalinity is given by ``Alk_C = \bar{Alk} + [B(OH)^-_4] + [OH^-]``, and we define the boron content, ``B``, to be ``R_{B:S}S`` mol/kg. | |||
| Finally, taking the DIC and Alkalinity in mol / kg and equivilants / kg (i.e. scaled by ``10^{-3}/\rho_o``) denoted by ``\bar{DIC}`` and ``\bar{Alk}``, the carbonate alkalinity is given by ``Alk_C = \bar{Alk} + [B(OH)^-_4] + [OH^-]``, and we define the boron content, ``B``, to be ``R_{B:S}S`` mol/kg. | |||
There was a problem hiding this comment.
spelling should be equivalents, I think
| # Centigrade to kelvin | ||
| T += 273.15 | ||
|
|
||
| # mili-equivilants / m³ to equivilants / kg |
There was a problem hiding this comment.
spelling should be "equivalents" throughout
|
@johnryantaylor do you think it would make sense to move carbon chemistry to be a property of Previously we had only used it in the CO2 flux but PISCES needs the calcite saturation for various things, and it makes sense to me that we separate it so it can be used for any model. It wouldn't cause many changes and will just be there to be called, e.g. by the boundary condition. |
|
Also, I will update the docs page at some point but the code now includes the pressure correction for the rate constants. I have added some tests which ensure we have the constants right (when using the "best practice" parameters) like how Kgen does, so I think the only difference between the functionality here and cbsyst on the carbon front is the Mg and Ca bits. |
|
Final comment for now, I'm pretty sure that this PR isn't breaking since |
…r things happen instead
|
TODO: check model_implemntation page when rendered update: it was broken |
validation/glodap_fit.jl
Outdated
| silicate = data[n, Si_name] * density * 1e-3 | ||
| phosphate = data[n, PO_name] * density * 1e-3 | ||
|
|
||
| silicate = ifelse(silicate == -9999, 0, silicate) |
There was a problem hiding this comment.
the ifelse statement should probably go before the units of silicate and phosphate are converted? Same with P
This PR refactors the carbon chemistry calculations improving the clarity and supersedes #179.
It also moves both the carbon chemistry and gas exchange functions to
Modelssince they are models, and separates them.This PR also implements the full carbon chemistry standards of the best practice (https://www.ncei.noaa.gov/access/ocean-carbon-acidification-data-system/oceans/Handbook_2007/Guide_all_in_one.pdf) so it can depend on Si and PO4, and approximated fluoride and sulfate. It also introduces proper (TEOS10 and polynomial approximations) density calculations.
Also, all of the equilibrium and pressure correction constants are also tested now.
I also introduce a function
calcite_saturationthat can computecalcite_concentrationand saturation for use in other models.This should also be faster since it removed a branching operation.
Finally, I changed the structure of
GasExchangeto make it more flexible (and correct, there was a bug in the computation of x(CO2) from fCO2).Also, closes #197, #200