Usage
TAfit(ae, titcurve, conc_titrant, mass_sample, S_titrant=NULL, TASumCO2guess=0.0025, E0guess=0.4, type="HCl", Evals=FALSE, electrode_polarity="pos", K_CO2fit=FALSE, equalspaced=TRUE, seawater_titrant=FALSE, pHscale="free", debug=FALSE, k_w=NULL, k_co2=NULL, k_hco3=NULL, k_boh3=NULL, k_hso4=NULL, k_hf=NULL, nlscontrol=nls.lm.control(), verbose=FALSE, k1k2="roy", khf="dickson", datxbegin=0, SumCO2Zero=FALSE)
Arguments
ae
an object of type aquaenv: minimal definition, contains all information about the system: T, S, d, total concentrations of nutrients etc (Note that it is possible to give values for SumBOH4, SumHSO4, and SumHF in the sample other than the ones calculated from salinity)
titcurve
a table containing the titration curve: basically a series of tuples of added titrant solution mass and pH values (pH on free proton scale) or E values in V
conc_titrant
concentration of the titrant solution in mol/kg-soln
mass_sample
the mass of the sample solution in kg
S_titrant
the salinity of the titrant solution, if not supplied it is assumed that the titrant solution has the same salinity as the sample solution
TASumCO2guess
a first guess for [TA] and [SumCO2] to be used as initial values for the optimization procedure
E0guess
first guess for E0 in V
type
the type of titrant: either "HCl" or "NaOH"
Evals
are the supplied datapoints pH or E (V) values?
electrode_polarity
either "pos" or "neg": how is the polarity of the Electrode: E = E0 -(RT/F)ln(H+) ("pos") or -E = E0 -(RT/F)ln(H+) ("neg")?
K_CO2fit
should K\_CO2 be fitted as well?
equalspaced
are the mass values of titcurve equally spaced?
seawater_titrant
is the titrant based on natural seawater? (does it contain SumBOH4, SumHSO4, and SumHF in the same proportions as seawater, i.e., correlated to S?); Note that you can only assume a seawater based titrant (i.e. SumBOH4, SumHSO4, and SumHF ~ S) or a water based titrant (i.e. SumBOH4, SumHSO4, and SumHF = 0). It is not possible to give values for SumBOH4, SumHSO4, and SumHF of the titrant.
pHscale
either "free", "total", "sws" or "nbs": if the titration curve contains pH data: on which scale is it measured?
debug
debug mode: the last simulated titration tit, the
converted pH profile calc, and the nls.lm output out are made global
variables for investigation and plotting
k_w
a fixed K\_W can be specified
k_co2
a fixed K\_CO2 can be specified; used for TA fitting: give a K\_CO2 and NOT calculate it from T and S: i.e. K\_CO2 can be fitted in the routine as well
k_hco3
a fixed K\_HCO3 can be specified
k_boh3
a fixed K\_BOH3 can be specified
k_hso4
a fixed K\_HSO4 can be specified
k_hf
a fixed K\_HF can be specified
nlscontrol
nls.lm.control() can be specified
verbose
verbose mode: show the traject of the fitting in a
plot
k1k2
either "roy" (default, Roy1993a) or "lueker" (Lueker2000) for K\_CO2 and K\_HCO3.
khf
either "dickson" (default, Dickson1979a) or "perez"
(Perez1987a) for K\_HF
datxbegin
at what x value (amount of titrant added) does the
supplied curve start? (i.e. is the complete curve supplied or just a
part?)
SumCO2Zero
should SumCO2==0?