-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathBOB_AMBROSE_mail.txt
141 lines (103 loc) · 6.06 KB
/
BOB_AMBROSE_mail.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
FOTOREAZIONI
fD <0.5 --> importante
cambia tantissimo anche in base ai decimali, conta piu che ref light o tasso trasf
Hi, Ginevra. First, I'm glad you are examining your output and asking these
questions. It is important to make sense of all the calculations. This is
the best way to discover errors or misunderstandings. Constructing mass
balances is most useful.
In the mercury model, the output rates are supposed to be in units of
[1/day]. The bacterial methylation reaction, I believe, only applies to the
solids-bound HgII in water and sediment. This is what you might be seeing
in the water column. If your input rate constant is 10^-3 and the sediment
bound fraction is 10^-5, then the effective transformation rate constant
would be 10^-8.
If I remember correctly, the water column also has photo-methylation and
demethylation reactions. What is your output interval? If it is set to the
default 1.0 days, then you might just be seeing the rates at midnight
(which should be 0 or some minimum value). If you set the output interval
to 0.1 or 0.05, you should see a nice diel curve for these reaction rates.
To judge whether the masses of mercury components seems correct, I'd have
to know your water column volumes. If total mercury is on the order of 1
ng/L (ug/m3), and your volume is on the order of 10^12 m3, then you should
have a tonne of total mercury (if I converted the units correctly).
That relationship between Kd and Hg0 is interesting, and I can't say I
understand it. Perhaps with higher Kd you are resuspending more HgII to the
water column where it partitions to the DOC phase and photo reduces. You
could test this by zeroing out resuspension and then checking to see
whether the relationship of Kd and HgII still holds true.
To help any more, I'd probably need to see your input file or some output
graphs. At present cannot run WASP on my new Mac. Maybe I could meet up
with Chris Knightes at the coffee shop if he has WASP on a laptop. We would
need to have your input file (*.wif) to look at. Alternatively, if you
could send your *.out file (which is ASCII), I might be able to diagnose
the problem looking at the data and messages.
#-----------
Ok, I've brushed up a bit on the model and the input data here. First, I
was (partly) wrong about the photo-transformations. This type of reaction
applies to photo-reduction of HgII to Hg0, and photo-demethylation of MeHg
to Hg0. If you plot those effective reaction rate constants, they will be 0
because your output is always at midnight. They will be quite small even
in the middle of the day because of the large light extinction coefficient
(which seems reasonable for the ultraviolet wavelengths).
Now, the methylation. As you point out, you are specifying a methylation
rate constant k_methyl of around 0.01 day^-1 applied only to the dissolved
fraction of HgII in the sediment compartments. The effective rate constant
will be
k_effective = k_methyl * f_diss * T_correct
The fraction dissolved will be quite small in the sediment compartments,
where the concentration of silt and organic solids are:
S_silt = 10^6 mg/L = 1.0 kg/L
S_org = 10^5 mg/L = 0.1 kg/L
The porosity, n, should be around 0.5 Lw/L.
The partition coefficients K_silt and K_Sorg are 2*10^3 and 2*10^5 Lw/kg
A rough calculation is:
f_diss = n / (n + K_silt * S_silt + K_Sorg * S_org) = 0.5 / (0.5 + 2*10^3 *
1.0 + 2*10^5 * 0.1) = 0.5 / 0.22*10^5 = 2 * 10^-5.
At 20C, the T_correct should be 1.0, and thus:
k_effective = 10^-02 * 2*10^-5 * 1 = 2 * 10^-7 day^-1
This matches up well with the lower concentrations in your plot of
methylation rates.
I don't have an explanation for how Hg0 concentrations (in the water
column, I presume) are correlated with increasing K_silt. It seems like
increasing K_silt will reduce the fraction of HgII on DOC in the water
column (which is maybe around 0.05 or so), and thus reduce the photo
reduction rate of HgII to Hg0 (since you have only the DOC fraction photo
reducing). If the increasing K_silt is somehow mobilizing more HgII from
the sediments to the water column, then there would be more photo
reduction. When you increase K_silt, how are the water column
concentrations of HgII and MeHg responding along with Hg0? It might also
help to plot the demethylation rate constants (you'd have to have output
during the day, which specifying a print interval of 0.1 would give you).
If you change the print interval to 0.1 for this test, you could run your
simulation for only a year to see what happens.
Bob
On Tue, Feb 24, 2015 at
So I see WASP has output for oxidation rate but not photo-oxidation rate of
Hg0. That was an oversight, and would not be hard to change. All the rates
are actually being calculated, but only two are output for each mercury
component. I don't know if Tim could recompile another mercury.dll for you.
Meanwhile, Ginevra, to calculate the effective photooxidation rate from the
effective photo reduction rate, you have to adjust for the phase fractions
(since your photo reduction is applied only to the DOC-complexed fraction
of HgII):
k_photooxidation = k_photoreduction * [k_photooxidation / k_photoreduction]* [(f_Hg0_diss + f_Hg0_DOC) / f_HgII_DOC]
f_Hg0_diss + f_Hg0_DOC = 1.0
f_HgII_DOC = K_DOC * DOC / [n + K_DOC * DOC + K_silt * S_silt + K_Sorg *
S_Sorg]
where DOC, S_silt, and S_org are expressed in units of kg/L.
For your calculations of input sediment concentration [mg_solids / L] based
on porosity, you need to use dry density, which has units of [kg_solids /
L] rather than bulk density, which has units of [kg_solids+water / L]. So
your total solids concentration in the surface and bottom layers would be
520,000 mg/L and 1,040,000 mg/L.
Bob
p.s., *Note to Tim Wool*: If you can recompile a WASP7 mercury.dll, here is
the change to TOXIDU in the mercury folder:
Change
DVAR(22,ISEG) = RATE(K,5) !oxid
to
DVAR(22,ISEG) = RATE(K,3) + RATE(K,5) !total oxidation
Precipitation/Evaporation Rate (M/SEC) for corresponding
DAY(K),HR(K),MIN(K). If rainfall units are cm/day or cm/year the conversion
factors are 1.157E-7 and 3.17E-10. NOTE: Precipitation is positive,
evaporation is negative.