quantifying and controlling error in salt dilution flow measurements gabe sentlinger (ard), andre...
TRANSCRIPT
AR s
Quantifying and Controlling Error in Salt Dilution Flow Measurements
Gabe Sentlinger (ARD), Andre Zimmermann (nhc),Mark Richardson (UBC), John Fraser (ARD)
AR s Quantifying Uncertainty
-We want to identify sources of error/uncertainty and control for them. -Ensure calibration, repeatability, linearity of method.
TCFdtbgECtEc
MassQ
.)()(
TCF
TCF
dt
dt
bgECtEC
bgECtEC
M
M
Q
Q
.
.
)()(
)()(
Components of Error
Lower Range
Upper Range Notes
Mixing 0% >200%Can be eliminated with long mixing length
Salt Mass <1% 10%Can be nearly eliminated with careful procedures
CF.T <1% 10% Temperature Compensation
BG EC.T <1% >40% Depending on SNR and EC.T Drift
Sensor Resolution <1% >40% Baseline and SNR
Summary <4% >300%
EC.T Uncertainty
AR s Salt Dilution Tracer Method
AR s Salt Dilution Tracer Method
Criteria [Cl] mg/l [NaCl] (mg/l) Peak over BGTaste 200 330 687
Benthic 800 1319 2747
ISO9555-3
-Previous studies have suggested a multiplier over background (ie 50% - 500%).-We recommend a SNR approach instead.-For Inst. Resolution of 0.1 uS/cm and low BG noise, a rise of 10 uS/cm is a SNR of 100. This introduces only 1% uncertainty into measurement.-Avoid Overdosing!
(from Moore, 2005)
Goals:1-Quantify Uncertainty2-Reduce Uncertainty and/or Dosage
Area/Doses*mg/l 10 30 60 10 30 60
200 5 2 1 10 3 2500 13 4 2 26 9 4
1,000 25 8 4 52 17 92,000 50 17 8 104 35 174,000 100 33 17 208 69 35
NOTESA) Peak : Average Factor 15
Peak mg/l for Transit Time (mins) Peak EC.T over BG EC.T for Transit Time (mins)
AR s
0
5
10
15
20
25
30
35
40
45
50
0 0.2
0.4
0.6
0.8
1 1.2
1.4
1.6
1.8
2 2.2
2.4
2.6
2.8
3 3.2
3.4
3.6
3.8
4
Dosage (kg/cms)
Pe
ak
ov
er
BG
EC
.T
60 mins
30 mins
10 mins
Peak over BG EC.T
-Typically we are aiming for a Peak over BG EC.T of 10-30 uS/cm, with at least a sensor resolution of 0.1uS/cm. If 1uS/cm resolution, then larger peak over BG EC.T is needed (larger dose). Larger dose needed for longer transit time. Inject large doses slooooowwwwwllllyyy.
AR s Duration of Injection not important
-Injection rate not important. Better to ensure it doesn’t sink to bottom and avoid WQ guideline problems at point of injection. Better to spread injection out.
-There is no significant difference between Pulse 1 (instantaneous) and Pulse 2 (slowly over 1min) or Pulse 1+Pulse 2 (sum of both masses).
90
95
100
105
110
115
120
125
130
135
140
10/17/14 7:53 10/17/14 7:56 10/17/14 7:59 10/17/14 8:02 10/17/14 8:05 10/17/14 8:08 10/17/14 8:11 10/17/14 8:13
DateTime
EC
.T(u
S/c
m @
25
C)
EC
(u
S/c
m)
EC.T
FINAL EC.T
Baseline
Start Baseline
End Baseline
S
E
Pulse 1
Q=0.085 m3/sUnc. =±3.7%
Pulse 2
Q=0.083 m3/sUnc. =±3.5%
Pulse 1 + 2
Q=0.0846 m3/sUnc. =±3.3%
AR s Lowest Fruit: Mixing Error
0.0
20.0
40.0
60.0
80.0
100.0
120.0
1/13/13 10:12 1/13/13 10:13 1/13/13 10:14 1/13/13 10:16 1/13/13 10:17 1/13/13 10:19 1/13/13 10:20 1/13/13 10:22
Date:Time
De
lta
[N
aC
l] m
g/m
l
Left Bank
Thalweg
Right Bank
• Difficult to quantify missing error. Multiple probes, or injections, can help determine appropriate mixing reach for given Q
• This is a single pulse measured on left and right bank, and thalweg. Standard Deviation ±21%.
Left Bank Q = 0.245 cmsThalweg Q = 0.297 cmsRight Bank Q = 0.345 cmsAverage Q = 0.304 cmsSwoffer Q = 0.297 cms
(from Carnation Creek, B.C. Courtesy of Robin Pike, MOE)
AR s Instrument Accuracy/Resolution
– What is the limit of the instrument?– This is a Unidata 6536B with resolution of 0.01 uS/cm but
accuracy of 0.5% of the reading. We found quanta of 0.4-0.6 uS/cm (+/-0.35%)
60
62
64
66
68
70
72
74
76
78
80
4/13/12 13:40 4/13/12 13:55 4/13/12 14:09 4/13/12 14:24 4/13/12 14:38 4/13/12 14:52 4/13/12 15:07 4/13/12 15:21 4/13/12 15:36Elapsed Time (s)
EC
.T@
25oC
(u
S/c
m)
EC
(u
S/c
m)
CCTBackground C (uS/cm/oC)Background CT (uS/cm)Alt_BaselineALT CT
1.3 kg/cms3.01 cms ± 11%
0.69 kg/cms3.68 cms ± 11%
0.29 kg/cms3.52 ± 16% 0.17 kg/cms
3.07 ± 104%
Unidata 6536BRes: 0.01 uS/cmAccuracy: 0.5%
step = 0.4-0.6 uS/cm = 0.7% = ± 0.35%
AR s Instrument Accuracy/Resolution
– Oakton Con110 has better accuracy and resolution at lower ranges, higher SNR, therefore less uncertainty in Q.
70
75
80
85
90
95
100
4/13/12 13:40 4/13/12 13:55 4/13/12 14:09 4/13/12 14:24 4/13/12 14:38 4/13/12 14:52 4/13/12 15:07 4/13/12 15:21 4/13/12 15:36Elapsed Time (s)
EC
.T@
25oC
(u
S/c
m)
EC
(u
S/c
m)
CCTBackground C (uS/cm/oC)Background CT (uS/cm)Alt_BaselineALT CT
1.33 kg/cms3.03 cms ± 4%
0.69 kg/cms2.90 cms ± 6%
0.29 kg/cms2.88 ± 6%
0.16 kg/cms2.96 ± 13%
Oakton CON110AutoRangingRes: 0.05% FullScale Accuracy: 1% FS + 1 digit0-20 uS/cm: 0.01 uS/cm, ±0.2 uS/cm20-200uS/cm: 0.1 uS/cm, ±2 uS/cm
step = 0.1 uS/cm
AR s
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 1000 2000 3000 4000 5000 6000 7000
Elapsed Time (s)
CT(u
S/c
m/o
C)
C (
uS
/cm
)
Measured
Modeled
Initial study examined SNR
– Difficult to assess true error without knowing Q.– Built an Instrument Model to model SNR. Random noise and Sensor
Resolution. Random noise is not divided between real and sensor. Modeled is simple chi-squared function.
– Noise in SNR includes quantization noise.– When Noise < Resolution, bias is likely, depending on background EC.T– Max Error is worst out of 20 simulations (α≈0.05)
0.0
5.0
10.0
15.0
20.0
25.0
0 1000 2000 3000 4000 5000 6000 7000
Elapsed Time (s)
CT(
uS
/cm
/oC
)
C (
uS
/cm
)
Measured
Modeled
Noise = 5, Res = 10, SNR = ~2, MaxE≈15% Noise = 1, Res = 1, SNR≈40, MaxE≈1.5%
Noise < Resolution = Bad
AR s
-2.0
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 1000 2000 3000 4000 5000 6000 7000
Elapsed Time (s)
CT(u
S/c
m/o
C)
C (
uS
/cm
)
Measured
Modeled
– Best Instrument for SD is• High Resolution and Quanta (0.1 uS/cm or better for EC.T <200uS/cm)• Repeatable, Linear, Stable (these are different than accuracy).• Absolute calibration not necessary if insitu CF.T derivation carried out
for each site.• Benefit of absolute calibration is that it gives expected CF.T values for
QA/QC.
-5.0
0.0
5.0
10.0
15.0
20.0
0 1000 2000 3000 4000 5000 6000 7000
Elapsed Time (s)
CT(u
S/c
m/o
C)
C (
uS
/cm
)
Measured
Modeled
Noise = 5, Res = 1, SNR ≈ 4, MaxE ≈ 13%
Noise > Resolution = Good
Noise = 1, Res = .1, SNR ≈ 90, MaxE ≈ 3%
Noise > Resolution = Good
Initial study examined SNR
AR s
Need for fast, accurate, reproducible method
– Reduce the amount of salt required for a given flow/ reduce uncertainty associated with salt dilution measurement.
– Establish SOP to ensure data quality and traceability, quantify uncertainty, and protect sensitive habitat.
Showing 95% Confidence (2s) results
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
11%
12%
13%
14%
15%
16%
17%
18%
19%
20%
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6
Salt Injection Concentration (kg/cms)
%U
nce
rta
inty
in
Q (
alp
ha
= 0
.05)
Inst.Err:0.01 uS/cm and Trans.Time: 10 mins
Inst.Err:0.01 uS/cm and Trans.Time: 30 mins
Inst.Err:0.01 uS/cm and Trans.Time: 60 mins
Inst.Err:0.1 uS/cm and Trans.Time: 10 mins
Inst.Err:0.1 uS/cm and Trans.Time: 30 mins
Inst.Err:0.1 uS/cm and Trans.Time: 60 mins
Inst.Err:1 uS/cm and Trans.Time: 10 mins
Inst.Err:1 uS/cm and Trans.Time: 30 mins
Inst.Err:1 uS/cm and Trans.Time: 60 mins
AQUARIUS RESEARCH & DEVELOPMENT INC.
AUTOSALT
Nov 2, 2014
Uncertainty as a function of Instrument Accuracy and Transit Time
Notes:1) This plot shows the relationship between instrument error and required salt slug injection concentration.2) The "Inst.Err." value is the uncertainty in the measurement, which is a function of the signal stability, the instrument resolution, and instrument accuracy.3) The current acceptable range of uncertainty and Injection concentration excludes instrument error of 1 uS/cm and transit time greater than 10 mins.4) The target range is primarily from an Instrument Error of 0.01 uS/cm. This resolution is not available on most commercial sensors when the baseline conductivity is greater than 20 uS/cm.5) Assumes complete mixing, stable background conductivity, and insignificant error from Concentration Factor and Salt Mass measurement. Uncertainty in background conductivity included.
Acceptable Range
Target Range
C:\Users\gsentlin\AQUARIUS_R&D\BUSINESS\Research&Development\Conferences\AWRA_Virginia\CF.T Calcs\[Raffuse_SD-CM_Apr 13 2012_InstrumentNoiseModel_v0.9.xls]Notes
AR s
– I generally use 10 points pre and 10 points post. I use SE of entire 20 point sample. This takes into account changing BG or incomplete trace.
– EC.T Unc. is larger of SE and ½ Sensor Resolution
Uncertainty in Area under Pulse
n
sSE TBGEC
TBGEC.
.
RSEUncTEC TBGEC 21,max. .
– Determine Standard Error (SE) of mean for pre- and post-.
– Assume SE is representative of signal and background error
n
sSE TBGEC
TBGEC.
.
AR s Uncertainty in Area under Pulse
– Noise = EC.TUnc*Trans.Time– Signal=Area under curve
=AverageEC.T*Trans.Time
– SNR=AverageEC.T/ECTUnc.– AreaUnc. = 1/SNR– Example:– Area under curve =1228
sec.uS/cm– Noise = 89.2 sec.uS/cm– SNR= 1228/89
=13.8 or 7.3% or 14.6% uncertainty at α=0.05
– *In most figures, ± 1σ is reported although most people think of error as ± 2σ.
AR s
-30%-25%-20%-15%-10%
-5%0%5%
10%15%20%25%30%35%40%45%50%55%60%
0 10 20 30 40 50 60 70 80 90
Index
De
lta
Q (
%)
Total
Delta Q vs Sqrt(Unc.12 + Unc.22)
-This figure shows the result of ~343 recent measurements.-87 Concurrent measurements.
#57#30
#61
AR s Discrepancy in #30
-#30 Mixing length too short.-DQ = 18.5%-Indicators: Shape of pulse not smooth.-DQ is larger than Unc1 + Unc2.-CM measurement was 3.3 m3/s.
6.0
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
5/20/147:55
5/20/148:02
5/20/148:09
5/20/148:16
5/20/148:24
5/20/148:31
5/20/148:38
5/20/148:45
5/20/148:52
5/20/149:00
5/20/149:07
5/20/149:14
DateTime
EC
.T(u
S/c
m @
25
C)
EC
(u
S/c
m)
0
2
4
6
8
10
12
14
Te
mp
era
ture
(oC
)
FINAL EC.T
Baseline
Start Baseline
End Baseline
S
E
Temp
Temp Spike removed from EC.T
Measurement stopped pre-maturelyEC.T Trace filled in. Q=4.9 m3/s
Unc. =±5.8%
6.0
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
5/19/14 7:55 5/19/14 8:02 5/19/14 8:09 5/19/14 8:16 5/19/14 8:24 5/19/14 8:31 5/19/14 8:38 5/19/14 8:45 5/19/14 8:52
DateTime
EC
.T(u
S/c
m @
25
C)
EC
(u
S/c
m)
0
2
4
6
8
10
12
14
Tem
per
atu
re (
oC
)
EC.T
FINAL EC.T
Baseline
Start Baseline
End Baseline
S
E
ALT EC
Temp
Measurement stopped pre-maturelyEC.T Trace wasn't filled in.
Much larger area
Q=4..14 m3/sUnc. =±4.3%
AR s Discrepancy in #57
-#57 Mixing length too short.-DQ = 52%-Indicators: Shape of LB pulse and-DQ is larger than Unc1 + Unc2.
90
91
92
93
94
95
96
97
98
99
100
9/12/1416:59
9/12/1417:00
9/12/1417:02
9/12/1417:03
9/12/1417:05
9/12/1417:06
9/12/1417:08
9/12/1417:09
9/12/1417:11
DateTime
EC
.T(u
S/c
m @
25C
)E
C (
uS
/cm
)
0
5
10
15
20
25
Tem
per
atu
re (
oC
)
FINAL EC.T
Baseline
Start Baseline
End Baseline
S
E
ALT EC
Temp
Center
Q=0.05 m3/sUnc. =±10.2%
100
101
102
103
104
105
106
107
108
109
110
9/12/1416:59
9/12/1417:00
9/12/1417:02
9/12/1417:03
9/12/1417:05
9/12/1417:06
9/12/1417:08
9/12/1417:09
9/12/1417:11
DateTime
EC
.T(u
S/c
m @
25C
)E
C (
uS
/cm
)
0
5
10
15
20
25
Tem
per
atu
re (
oC
)
FINAL EC.T
Baseline
Start Baseline
End Baseline
S
E
ALT EC
Temp
LB
Q=0.10 m3/sUnc. =±14.7%
AR s Discrepancy in #61
-#61 Large temperature compensation, temperature change, dosage too low.
-DQ = 18%, 1.1 kg/m3/s.-Indicators: BG ECT not steady-CM measurement was 10.2 m3/s.
70
80
90
100
110
120
130
9/16/14 12:00 9/16/14 12:14 9/16/14 12:28 9/16/14 12:43 9/16/14 12:57 9/16/14 13:12 9/16/14 13:26DateTime
EC
.T(u
S/c
m @
25
C)
EC
(u
S/c
m)
5
5.5
6
6.5
7
7.5
8
Te
mp
era
ture
(oC
)
QM4.0_TM3.1_ECT
QM4.0_TM2.5_ECT
Baseline
Start Baseline
End Baseline
S
E
QM4.0_TM2.5_EC
QM4.0_TM2.5_Temp
TM3.1 on RB
Q = 10.5 m3/s%Unc. = 4.0%
TM2.5 on LB
Q = 8.9 m3/s%Unc. = 4.6%
AR s Experimental Setup
-Typical setup with one probe mid-channel upstream, second probe left bank downstream.
-DQ and %Unc. Calculated in realtime to aid in decision making.
T-HRECS on CH1
QuickFlask
DuckBox
QiQuac
T-HRECS on CH0
QuickFlock on CH1
100
105
110
115
120
125
10/15/14 9:46 10/15/14 9:48 10/15/14 9:51 10/15/14 9:54 10/15/14 9:57 10/15/14 10:00 10/15/14 10:03 10/15/14 10:06 10/15/14 10:09
DateTime
EC
.T(u
S/c
m @
25
C)
EC
(u
S/c
m)
US-MID
DS LB
Baseline
Start Baseline
End Baseline
S
EDS-LB Q= 0.078±7%
US-MID Q= 0.087±7%
US-MID Q= 0.090±13%
DS-LB Q= 0.093±15%
-US-MID higher peak, earlier, but approximately the same area (no significant difference between derived Q at 95% confidence)
AR s
Total SDIQ %Uncertainty vs Dosage vs EC.T Unc.
0%
10%
20%
30%
40%
50%
60%
70%
0 1 2 3 4
Dosage (kg/cms)
To
tal
%U
nc
ert
ain
ty
0.02 µS/cm
0.1 µS/cm
1 µS/cm
Total % Uncertainty
-This figure shows the result of 343 recent measurements (2013-2014).-Lowest Uncertainty when EC.T Unc. is smallest.-Baseline of ~3% Uncertainty from CF.T and NaCl Mass.
Total <10% <5% <3%343 270 57 4
100% 79% 17% 1%
AR s
%Uncertainty due to EC.T Unc.
0%
5%
10%
15%
20%
25%
30%
0 0.2
0.4
0.6
0.8
1 1.2
1.4
1.6
1.8
2 2.2
2.4
2.6
2.8
3 3.2
3.4
3.6
3.8
4
Dosage (kg/cms)
%U
nc
ert
ain
ty
0.02 µS/cm
0.1 µS/cm
1 µS/cm
% Uncertainty due to EC.T Unc.
-Remove Unc. Due to CF.T and Mass.-Unc. Due to SNR steady at 5% until ~0.7 kg/cms
Total <10% <5% <3%343 303 275 225
100% 88% 80% 66%
AR s
EC.T.Uncertainty = 0.1 uS/cm
0%
5%
10%
15%
20%
25%
30%
0 0.2
0.4
0.6
0.8
1 1.2
1.4
1.6
1.8
2 2.2
2.4
2.6
2.8
3 3.2
3.4
3.6
3.8
4
Dosage (kg/cms)
%U
nc
ert
ain
ty
60 mins
30 mins
10 mins
Target Range
-We can further explain variance by transit time of pulse.-The faster the transit time, assuming completely mixed, the lower the
uncertainty.-0.1 µS/cm typical of many commercial sensors for EC.T < 200 µS/cm.
% Uncertainty due to EC.T Unc.:0.1µS/cm
Total <10% <5% <3%185 164 146 115
100% 89% 79% 62%
0%
5%
10%
15%
20%
25%
30%
0 0.2
0.4
0.6
0.8
1 1.2
1.4
1.6
1.8
2 2.2
2.4
2.6
2.8
3 3.2
3.4
3.6
3.8
4
AR s
Inst.Uncertainty = 0.1 uS/cm
0%
5%
10%
15%
20%
25%
30%
0 0.2
0.4
0.6
0.8
1 1.2
1.4
1.6
1.8
2 2.2
2.4
2.6
2.8
3 3.2
3.4
3.6
3.8
4
Dosage (kg/cms)
%U
nc
ert
ain
ty
60 mins
30 mins
10 mins
Inst.Err:0.1 uS/cm and Trans.Time: 60 mins
Inst.Err:0.1 uS/cm and Trans.Time: 30 mins
Inst.Err:0.1 uS/cm and Trans.Time: 10 mins
Target Range
I compared to my 2012 estimates of Uncertainty (x0.5 to get 1σ). Good agreement with 0.1 µS/cm above 1 kg/cms, but overestimated for less than 1 kg/cms.
% Uncertainty due to EC.T Unc.:0.1µS/cm
AR s
EC.T.Uncertainty= 0.02uS/cm
0%
5%
10%
15%
20%
25%
30%
0 0.2
0.4
0.6
0.8
1 1.2
1.4
1.6
1.8
2 2.2
2.4
2.6
2.8
3 3.2
3.4
3.6
3.8
4
Dosage (kg/cms)
%U
nc
ert
ain
ty
60 mins
30 mins
10 mins
Target Range
% Uncertainty due to EC.T Unc.: 0.02 µS/cm
-If we can attain EC.T Unc. of 0.02 µS/cm (requires finer than 0.02 µS/cm resolution), then we can reduce the dose to 0.2 kg/cms and remain below 5% Uncertainty.
Total <10% <5% <3%114 114 110 99
100% 100% 96% 87%
AR s
EC.T.Uncertainty= 0.02uS/cm
0%
5%
10%
15%
20%
25%
30%
0 0.2
0.4
0.6
0.8
1 1.2
1.4
1.6
1.8
2 2.2
2.4
2.6
2.8
3 3.2
3.4
3.6
3.8
4
Dosage (kg/cms)
%U
nc
ert
ain
ty
60 mins
30 mins
10 mins
Inst.Err:0.02 uS/cm and Trans.Time: 10 mins
Inst.Err:0.02 uS/cm and Trans.Time: 30 mins
Inst.Err:0.02 uS/cm and Trans.Time: 60 mins
Target Range
% Uncertainty due to EC.T Unc.: 0.02 µS/cm
- Good agreement with 0.02 µS/cm.- Used 0.02 µS/cm because Inst.Res 0.01 µS/cm but still shows
noise, partly due to temperature.
AR s
-EC.T Unc. Ceiling 1 µS/cm or greater. Combined sensor resolution, background noise, and return to background EC.T.
-Only two measurements in target range.-Possible to get high quality measurement, but requires larger dose.
EC.T. Uncertainty = 1uS/cm
0%
5%
10%
15%
20%
25%
30%
0 0.2
0.4
0.6
0.8
1 1.2
1.4
1.6
1.8
2 2.2
2.4
2.6
2.8
3 3.2
3.4
3.6
3.8
4
Dosage (kg/cms)
%U
nc
ert
ain
ty
60 mins
30 mins
10 mins
Target Range
% Uncertainty due to EC.T Unc.: 1µS/cm
Total <10% <5% <3%44 25 19 11
100% 57% 43% 25%
AR s
EC.T. Uncertainty = 1uS/cm
0%
5%
10%
15%
20%
25%
30%
0 0.2
0.4
0.6
0.8
1 1.2
1.4
1.6
1.8
2 2.2
2.4
2.6
2.8
3 3.2
3.4
3.6
3.8
4
Dosage (kg/cms)
%U
nc
ert
ain
ty
60 mins
30 mins
10 mins
Inst.Err:1 uS/cm and Trans.Time: 30 mins
Inst.Err:1 uS/cm and Trans.Time: 10 mins
Target Range
% Uncertainty due to EC.T Unc.: 1µS/cm
- Unc. Overestimated for EC.T Unc. = 1 us/cm. Not sure why yet
AR s Recommendations
– The relationship shown in previous figures is tabulated here for two data classes. NaCl dose depends on EC.T Unc. and transit time.
– Most common EC.T Unc. and transit time (0.1 uS/cm and 30 mins) requires 1.1 kg/cms to attain ±5% at 95%confidence.
– Shorter transit time or lower EC.T Unc. results in lower dose required.– Rule of Thumb:
• 10x the EC.T Unc. at 30 mins. Except at 1uS/cm use 2kg/cms.• Half the dose for 10 mins, double for 60 mins.
– Best Practice: Pre-dissolve or inject slowly (>30 second).
C:\Users\gsentlin\AQUARIUS_R&D\BUSINESS\Research&Development\Conferences\AWRA_Virginia\CF.T Calcs\[Raffuse_SD-CM_Apr 13 2012_InstrumentNoiseModel_v0.9.xls]SNR_%Error 10/31/14
(2.5% SE) (5.0% SE)EC.T Unc. (µS/cm) 10 30 60 EC.T Unc. (µS/cm) 10 30 60
0.01 0.04 0.11 0.21 0.01 0.02 0.06 0.130.02 0.07 0.21 0.42 0.02 0.04 0.13 0.250.1 0.35 1.1 2.1 0.1 0.2 0.6 1.31 1.2 2 5 1 0.8 1.2 6
NOTES
B) Dose estimates from the EC.T Unc. = 1 uS/cm are estimated visually from scatter plots.
VER 0.6
A) EC.T Unc. Is the total uncertainty in the measurement due to Instrument Resolution, BG Noise, drifting background (due to temperature or EC).
Transit Time (mins)Transit Time (mins)Dose (kg/cms) Required for 5% Uncertainty at 95% Dose (kg/cms) Required for 10% Uncertainty at 95%
AR s Acknowledgements
T-HRECS on CH1
QuickFlask
DuckBox
QiQuac
T-HRECS on CH0
QuickFlock on CH1
100
105
110
115
120
125
10/15/14 9:46 10/15/14 9:48 10/15/14 9:51 10/15/14 9:54 10/15/14 9:57 10/15/14 10:00 10/15/14 10:03 10/15/14 10:06 10/15/14 10:09
DateTime
EC
.T(u
S/c
m @
25
C)
EC
(u
S/c
m)
US-MID
DS LB
Baseline
Start Baseline
End Baseline
S
EDS-LB Q= 0.078±7%
US-MID Q= 0.087±7%
US-MID Q= 0.090±13%
DS-LB Q= 0.093±15%
Thank you
Jane Bachman, EDI YukonDave Hutchinson, WSCRobin Pike, BC MOEDan Moore, UBC
Questions?