&EPA
United States
Environmental Protection
Agency
600R04053
Rapid Processing of Turner
Designs Model 10-AU-005
Internally Logged Fluorescence
Data
-------
Cover Figure
Flow chart for FLOWTHRU program subroutine TIMCNT explaining how FLOWTHRU handles
severe time breaks in logged data files.
-------
EPA/600/R-04/053
July 2004
Rapid Processing of Tlirner Designs
Model 10-AU-005 Internally
Logged Fluorescence Data
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection Agency
policy and approved for publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
ABSTRACT
Continuous recording of dye fluorescence using field fluorometers at selected sampling sites
facilitates acquisition of real-time dye tracing data. The Turner Designs Model 10-AU-005 field
fluorometer allows for frequent fluorescence readings, data logging, and easy downloading to
a laptop computer. By necessity, the data are periodically broken up into blocks to facilitate
downloading and minimize data loss. Unfortunately, the downloaded data do not appear in a
readily usable form for discerning trends or for use in modeling packages. A new computer
program, FLOWTHRU, bypasses block headers, reads the downloaded data, identifies the time-
concentration units used, and relates the data to injection time. All preinjection time-concentration
data are accorded background data status and are written to a background file with average
temperature values included. All time values recorded after injection time are rewritten into
decimal time in time units chosen by the user. Additional features include options for processing
selected percentages of the time-concentration data and for using the measured concentrations,
average concentrations of data blocks, or smoothed concentrations developed using a moving
average filter. FLOWTHRU also allows users to view the data converted to decimal time directly on
the computer monitor without program interruption or to go directly to the data plotting routine.
Data plotting is extremely rapid and clear, with a smooth line connecting each data point. Each
data plot may be saved as a file in a common format.
Preferred citation:
U.S. Environmental Protection Agency (EPA). (2004) Rapid processing of turner designs model 10-AU-005 internally
logged fluorescence data. National Center for Environmental Assessment, Washington, DC; EPA/600/R-04/053.
Available from: National Technical Information Service, Springfield, VA and http: / /www. epa. gov/ncea.
11
-------
Contents
ABSTRACT "
LIST OF TABLES v
LIST OF FIGURES vi
NOTATION viii
DESCRIPTION OF UNITS ix
PREFACE xi
AUTHOR and REVIEWERS xii
1. INTRODUCTION 1
1.1. MODEL 10-AU-005 DATA FILE APPEARANCE 2
1.1.1. Data Blocks 2
1.1.2. Data Interrupts 4
1.1.3. Large Data Sets 4
1.2. DESIRED IMPROVEMENTS 4
1.2.1. Form of Data File 4
1.2.2. When to Start Counting 5
1.2.3. Data Output 5
1.2.4. Data Plotting 5
2. MAIN PROGRAM DESIGN 6
2.1. MAIN PROGRAM 6
2.1.1. Background Concentration 6
2.1.2. Measured Water Temperature 10
2.1.3. Decimal Time Units 12
2.1.3.1. Decimal Time Calculation 12
2.1.4. Data Counting 13
2.1.5. Time Data Processing 14
2.1.6. Data Plotting 14
2.1.6.1. PostScript File Descriptions 14
2.1.7. Memory Usage 15
2.1.7.1. Dynamic Memory Allocation 17
3. PROGRAM SUBROUTINES 18
3.1. INTERPROCESS CONTROL SUBROUTINE 18
3.1.1. Line-by-Line Description of Input Files 23
3.2. ARRAY-SIZE ALLOCATION ADJUSTMENT SUBROUTINE 30
3.2.1. Data-Counting Control 32
in
-------
3.2.2. Data Time Breaks 32
3.2.2.1. Leap Year Calculation 32
3.3. DECIMAL TIME FILE AND PLOT FILE HEADERS '.'.'.'. 35
3.4. TIME-CONCENTRATION DATA AVERAGING 35
3.5. MEAN TEMPERATURE CALCULATION 36
4. USING FLOWTHRU TO EVALUATE MODEL 10-AU-005 DATA 37
4.1. FLOWTHRU PROGRAM USAGE AND EXAMPLE DATA FILES 37
4.1.1. Loading FLOWTHRU and Example Data Files 37
4.2. FLOWTHRU EXECUTION \ 38
4.3. COMPUTER GRAPHICS ' ' ' 40
4.3.1. Features of the Interactive Graphics Loop 40
4.3.1.1. Interactive Data Point Deletion 41
4.4. FLOWTHRU SOURCE '.'.'.'.'.'. 43
5. EXAMPLE RESULTS 47
5.1. PROCESSING SIMPLE DATA FILES 47
5.1.1. Turner Designs Data File TracqOl.prn 47
5.1.2. Turner Designs Data File TracqOl.prn 48
5.1.3. Moving Average Filter Effect on a Data File 51
5.1.3.1. Absolute and Relative Smoothing Errors 53
5.1.3.2. FLOWTHRU Analysis of the Measured and Modified Data Sets. 53
5.1.3.3. CXTFIT2 Analysis of the Measured and Modified Data Sets. . . 59
5.2. PROCESSING MODERATELY COMPLEX DATA FILES 59
5.2.1. Turner Designs Data File Creek.prn 68
5.3. PROCESSING VERY COMPLEX DATA FILES '.'.'.'.'.'. 68
5.3.1. Turner Designs Data File Fluor_time.prn 70
5.3.2. Turner Designs Data File Break.prn 74
5.3.3. Turner Designs Data File 03gl58.prn 74
5.3.3.1. Effect of Data Point Deletion on File 03glSS.prn 76
6. SUMMARY 80
APPENDIX I 81
REFERENCES 95
IV
-------
List of Tables
1 Internal data storage capacity of a Turner Designs Model 10-AU-005 1
2 Allowable Turner Designs Model 10-AU-005 concentration units 9
3 Multiplier for decimal time units 13
4 Description of the input file components listed in Figure 9 21
5 Pull-down menu items available in FLOWTHRU 44
6 Evaluation of measured and modified concent, parameters for TracqOl.prn .... 57
7 Evaluation of measured and modified transport parameters for TracqOl.prn. ... 57
8 Evaluation of measured and modified concent, parameters for Tracq04.prn .... 58
9 Evaluation of measured and modified transport parameters for Tracq04.prn. ... 58
10 Evaluation of equil. model estimated transport parameters for TracqOl.prn .... 66
11 Evaluation of equil. model estimated transport parameters for Tracq04.prn .... 66
12 Evaluation of nonequil. model estimated transport parameters for TracqOl.prn . . 67
13 Evaluation of nonequil. model estimated transport parameters for Tracq04.prn . . 67
-------
List of Figures
1 Example data file downloaded from a Turner Designs Model 10-AU-005 field
fluorometer 3
2 Flowchart depicting processing of the program FLOWTHRU 7
3 Example background data file produced by FLOWTHRU 10
4 Equations used to calculate univariate statistics 11
5 Univariate statistics created for a Turner Designs data file 11
6 Example water temperature data file produced by FLOWTHRU 12
7 Three PostScript formats produced by FLOWTHRU 16
8 Flowchart depicting subroutine for implementing interprocess control 19
9 Generic example of interprocess control input file format 20
10 Flowchart depicting subroutine for memory size using seconds 31
11 Flowchart depicting subroutine for counting past blocks 32
12 Flowchart depicting subroutine for counting past time breaks 33
13 Flowchart depicting subroutine for leap year 34
14 Initial FLOWTHRU screen title 39
15 Decimal time-concentration data file for the TracqOl.prn data file 49
16 Decimal time-concentration data file for the TracqOl.prn data file 49
17 Decimal time-concentration data file for the Tracq04.prn data file 50
18 Decimal time-concentration data file for the Tracq04.prn data file 50
19 Smoothed decimal time-concentration data file for the TracqOl.prn data file ... 52
20 Smoothed decimal time-concentration data file for the Tracq04.prn data file ... 52
21 Comparison of measured and smoothed concentrations TracqOl.prn data file . . . 54
22 Comparison of measured and smoothed concentrations Tracq04.prn data file ... 54
23 Plot of absolute errors between measured and smoothed concentrations for the
TracqOl.prn data file 55
24 Plot of relative errors between measured and smoothed concentrations for the
TracqOl.prn data file 55
25 Plot of absolute errors between measured and smoothed concentrations for the
Tracq04.prn data file 56
26 Plot of relative errors between measured and smoothed concentrations for the
Tracq04.prn data file 56
27 Plot of equilibrium model fit to the measured TracqOl.prn data 60
28 Plot of equilibrium model fit to the averaged TracqOl.prn data 60
29 Plot of equilibrium model fit to the smoothed TracqOl.prn data 61
30 Plot of nonequilibrium model fit to the measured TracqOl.prn data 61
31 Plot of nonequilibrium model fit to the averaged TracqOl.prn data 62
32 Plot of nonequilibrium model fit to the smoothed TracqOl.prn data 62
33 Plot of equilibrium model fit to the measured Tracq04.prn data 63
34 Plot of equilibrium model fit to the averaged Tracq04.prn data 63
35 Plot of equilibrium model fit to the smoothed Tracq04.prn data 64
36 Plot of nonequilibrium model fit to the measured Tracq04.prn data 64
37 Plot of nonequilibrium model fit to the averaged Tracq04.prn data 65
vi
-------
38 Plot of nonequilibrium model fit to the smoothed Tracq04.prn data 65
39 Decimal time-concentration data file for the Creek.prn data file 69
40 Decimal time-concentration data file for the Creek.prn data file 69
41 Decimal time-concentration data file for the Fluor Jime.prn data file 71
42 Decimal time-concentration data file for the Fluor Jime.prn data file (daily) ... 71
43 Decimal time-concentration data file for the FluorJlme.prn data file (every other
day) 72
44 Decimal time-concentration data file for the Fluor Jime.prn data file (every fifth day) 72
45 Decimal time-concentration data file for the Fluor Jime.prn data file (every tenth
day) 73
46 Decimal time-concentration data file for the Fluor Jime.prn data file (2% of data) 73
47 Decimal time-concentration data file for the Break.prn data file 75
48 Decimal time-concentration data file for the Break.prn data file 75
49 Decimal time-concentration data file for the 03gl58.prn data file (100% of data) . 77
50 Decimal time-concentration data file for the 03gl58.prn data file (50% of data) . 77
51 Decimal time-concentration data file for the 03gl58.prn data file (11% of data) . 78
52 Decimal time-concentration data file with two deleted data points 78
vn
-------
NOTATION
Cb Background concentration [M L~3]
Cb Average background concentration [M L~3]
Cp Peak concentration [M L~3]
d Calculated number of days from time of injection
Dx Axial dispersion [L2 T"1]
L Solute migration distance [L]
n Total number of time values to be recalculated to decimal time
rib Number of background measurements
Pe Peclet number [dimen.]
Q System discharge [L3 T"1 ]
As, Time spacing for exact or even time (1 < As, < 10 for As, in days and 24 < As, < 240
for As, in hours)
t Decimal time calculated from measured time using a 24-hour clock [T]
th Measured time in hours [h]
thr Tracer release time hour [h]
tmr Tracer release time minutes [min]
tm Measured time in minutes [min]
tp Time to peak arrival [T]
ts Measured time in seconds [s]
tst Calculated tracer injection (start) time in hours [h]
tu Multiplier to adjust decimal hours to decimal time units of choice
vm
-------
DESCRIPTION OF UNITS
Unit of Measure
Type of Unit
Abbreviation
Concentration Units
None
full scale units
quantitative fluorescent technique
parts per million
parts per billion
parts per trillion
milligrams per liter
milligrams per deciliter
grams per liter
milligrams per killigram
milligrams per gram
micrograms per liter
micrograms per milliliter
micrograms per kilogram
micrograms per gram
nanograms per liter
nanograms per milliliter
nanograms per microliter
nanograms per kilogram
nanograms per gram
nanograms per milligram
picograms per gram
picograms per milligram
picograms per microgram
picograms per milliliter
picograms per microliter
femtograms per milliliter
femtograms per milligram
relative concentration
fsu
qft
ppm
ppb
ppt
1
mgdL
mgkg
mgg"1
_
(jigmL !
M-g kg"1
ng mL
ng
ng kg"1
ngg"1
ng mg"1
Pg g-1_
Pg mg~!
Pg M'g^1
pgmL"1
Pg
fg
fgmg l
Raw
microgram
Mass Units
M-g
continued on next page
IX
-------
Unit of Measure Type of Unit Abbreviation
Time Units
days d
hours h
reciprocal hours h~i
minutes min
seconds s
Transport Units
meters m
kilometers km
meters per second m s-i
square meters per second m2 s-i
volumetric discharge m3 ^-i
Temperature Units
Celsius
Fahrenheit
Generic Units
length
mass
time
-------
PREFACE
The National Center for Environmental Assessment has prepared this document for the benefit of
the U.S. Environmental Protection Agency regional offices, states, and the general public because
of the need to develop a fast and easy method for rewriting data internally logged by a Turner
Designs Model 10-AU-005 in decimal time. Application of the methodology described in this
document can provide individuals with the information necessary for rapidly examining tracer
test results and preparing data sets for numerical analysis.
The purpose of this document and program is to serve as a technical guide to various groups
who must evaluate:
1. Travel times;
2. Dispersivities;
3. Hydraulic connections; and
4. Contributing (recharge) areas.
Tracing studies are always appropriate and probably necessary, but internally logged data are
not always directly usable. This document and associated computer programs alleviate some of
these problems.
The program, FLOWTHRU, allows for very rapid and easy rewriting of logged time into decimal
time, prepares selected data files, and plots the data directly so that data trends may be discerned.
It is believed that the use of FLOWTHRU will substantially enhance tracer test analyses.
This document and program are specific to the Turner Designs Model 10-AU-005 because
it is the instrument currently used by the EPA author. EPA does not endorse this instrument or
any other instrument of similar functionality. The original purpose behind the development of
FLOWTHRU was to enhance data manipulation of the logged data for the equipment in use. Once
developed, it was deemed sufficiently useful such that it should be placed in the public domain.
Although specific to the Turner Designs Model 10-AU-005, FLOWTHRU may be easily adapted to
other fluorometers with flowthrough-analysis capabilities provided a method is readily available
for direct calculation of tracer dye concentrations.
XI
-------
AUTHORS AND REVIEWERS
The National Center for Environmental Assessment within the U.S. Environmental Protection
Agency's Office of Research and Development was responsible for preparation this document and
provided overall direction and coordination during the production effort.
AUTHOR
Malcolm S. Field, Ph.D.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
REVIEWERS
Nicolas Massei, Ph.D.
Departmente of Geologic
Universite de Rouen
UMRCNRS6143 M2C
10, Boulevard de Broglie
76821 Mont-Saint-Aignan cedex
France
Michael Verreault, ing., M.Sc.A.
Les Laboratoires SL
1309, blv St-Paul
Chicoutimi, Quebec
Canada
Lawrence E. Spangler, M.S.
U.S. Geological Survey
Utah District Office
2329 Orton Circle (2300 South)
Salt Lake City, UT 84119-2047
XII
-------
1. INTRODUCTION
Continuous data recording at selected sampling sites is widely recognized as extremely
valuable for inexpensively obtaining large amounts of real-time data. Water level measurements
in piezometers using pressure transducers and external data loggers are a common example of
large amounts of data being logged for later analysis. Automatic fluorometric data logging of
tracer dyes using field fluorometers (Turner, 1993) and external data loggers (Smart et al., 1998)
are also common (Lin et al., 2003; Wilson et al., 1986, pp. 17-18).
The older Turner Model 10 Series fluorometers are still widely in use, but the Turner
Designs Model 10-AU-005 benefits from a number of improvements (Smart et al., 1998). The
Model 10-AU-005 field fluorometer, when fitted with a one-piece flow cell (Turner, 1993),
allows fluorescence logging from every second to every 30 minutes with automatic temperature
compensation included. However, the user is limited to selecting logging intervals of 1, 2, 3, 5,
10, 20, or 30 seconds or 1, 2, 3, 5, 10, 20, or 30 minutes. The total number of data points that
can be logged is also limited by its memory capability (Table 1). Although a very large number of
Table 1. Internal data storage capacity of a Turner Designs
Model 10-AU-005.
Logging Strategy3
Cycle
Cycle
One way
One way
Temperature Logging
With Temperature
Without Temperature
With Temperature
Without Temperature
Storage Sizeb
18,510
21,600
43,200
64,800
a Method of logging data. Cycle = Data logged chronologically
until the memory is full, then old data are overwritten starting
with oldest data. One way = Data logged chronologically until
the memory is full, then no more data may be logged.
b Total number of data values that may be logged.
data values may be logged by the Model 10-AU-005, it is still fairly limited relative to an external
data logger.
The Turner Designs IDL_1B 1 program (Turner, 1993) provides for relatively easy downloading
of data to a laptop computer, although the program currently must be run in the MSDOS mode1.
However, by necessity, the actual logged data do not appear in a form that is amenable to data
'Turner Designs has recently developed a Windows version of the !DL_lBl program which may be obtained at
http://www.turnerdesigns.com/t2/sw/main.html.
1
-------
analysis and plotting. Typically, users of data downloaded from a Model 10-AU-005 must use a
wordprocessor to remove block headers, edit the data, save the changes, and export the edited file
into a spreadsheet program for manipulation (Turner, 1993). For large time-concentration files,
this procedure can be onerous.
The purpose of this paper is to present a new computer program, FLOWTHRU, that facilitates
manipulation of Model 10-AU-005 data downloaded using the !DL_lBl program with minimal
effort by the user. FLOWTHRU is not intended as a replacement for the iDL.lBl program.
Rather, FLOWTHRU provides for easy and rapid data processing and plotting of Model 10-AU-
005 logged data without requiring the use of several different programs, file storage forms, or
plotting software.
1.1. MODEL 10-AU-005 DATA FILE APPEARANCE
A typical logged data file downloaded from a Model 10-AU-005 fluorometer is shown in Figure 1.
Several points can be made using this figure. Every logged data file begins with a Turner Designs
header file that is then followed by a brief description of the fluorometer setup. The actual logged
data are then listed sequentially. No column labels are provided, but the data appears as listed in
Figure 1 in columns organized as:
Number Date Time Concentration Temperature Temperature Units
where Temperature and Temperature Units do not appear if temperature logging is not set
at the time of startup. The appearance of the data is relatively self-explanatory. Apparent
in Figure 1 are the negative concentrations that can—and—often are recorded as a result of
background fluorescence in the water sample used to "blank" or zero the fluorometer. All negative
concentrations need to be converted to zero concentration, which is done by some model packages
(e.g., QTRACER2). Negative concentrations may be avoided by calibrating with distilled water as
a true blank then prior to any numerical analysis, reading the high blanks and subtracting them
from the other readings (Turner, 1993).
1.1.1. Data Blocks
By necessity, the data file is periodically broken up into blocks to facilitate downloading. Also
apparent from Figure 1 is that as new data blocks are downloaded, sample numbering begins anew
for each new data block. This renumbering complicates search-and-replace operations using the
sample numbers as pointers when editing the data.
-------
* TURNER DESIGNS
*
FLUOROMETER INTERNAL DATA LOGGER
DATA OUTPUT PROGRAM
DATA LOGGING STRATEGY: ONE WAY
DATA SET NO: 1
DATA LOGGING METHOD: INSTANT
DATA LOGGING INTERVAL: 10 (MIN)
DATA LOGGING UNIT: (ug/1)
DATA LOGGING STARTED: 10/11/02 10:24:50
DATA LOGGING ENDED: 10/18/02 10:14:50
00001: 10/11/02 10:24:50 = -1.20 14.0 (C)
00002: 10/11/02 10:34:50 = -1.15 14.0 (C)
01007: 10/18/02 10:04:50 = -0.907 13.7 (C)
01008: 10/18/02 10:14:50 = -0.917 13.7 (C)
DATA SET NO: 2
DATA LOGGING METHOD: INSTANT
DATA LOGGING INTERVAL: 10 (MIN)
DATA LOGGING UNIT: (ug/1)
DATA LOGGING STARTED: 10/18/02 10:41:36
DATA LOGGING ENDED: 11/05/02 15:25:48
00001: 10/18/02 10:41:36 = -0.931 13.7 (C)
00002: 10/18/02 10:51:36 = -0.966 13.7 (C)
00606: 11/05/02 15:15:48 = 0.072 13.7 (C)
00607: 11/05/02 15:25:48 = 0.028 13.7 (C)
Figure 1. Example data file downloaded from a Turner Designs Model 10-AU-005 field fluoro-
meter.
-------
Less apparent in Figure 1 is the fact that a break in the set time sequence (i.e., data-logging
interval = 10 min in Figure 1) can occur during data downloading (e.g., time intervals > 10 min).
This time break may occur because use of the iDL_lfil program requires that the Model 10-AU-
005 internal logger be shut down to avoid data loss during the downloading process. However, it
is possible that a scheduled sample measurement may coincide with logger shut down, causing a
minor loss of data. Although the loss of one or more sample readings may not be significant, such
a loss hampers the ability to perform a search-and-replace operation using measured time values
as pointers when editing the data.
1.1.2. Data Interrupts
Power surges or loss of power also may result in large data interrupts, or worse, corrupted data.
The only way to handle corrupted data is to erase all previously logged data and restart the logger.
The newly logged data can either be treated as a stand-alone data file or they can be appended to
a previously downloaded data file. In either instance, the resulting file is much more difficult to
evaluate than if no data corruption had occurred.
1.1.3. Large Data Sets
Lastly, depending on the selected logging interval and the length of time that the fluorometer is
run, very large data sets may be recorded. Data records »10,000 (but not exceeding 64,800 lines)
are not unrealistic but are difficult to analyze because of the large number of data values.
1.2. DESIRED IMPROVEMENTS
To represent an improvement over the existing Model 10-AU-005 data file and the wordproces-
sor/spreadsheet analysis procedure, a new program should be capable of directly reading, process-
ing, and rewriting the data file into a more useful form. All data obstructions (e.g., block headers)
should be seamlessly bypassed. Manual interaction by the user should be limited to providing
certain desirable aspects specified by the user.
1.2.1. Form of Data File
Although appropriate as listed, the data file shown in Figure 1 is difficult to analyze, and data
trends are difficult to discern. These problem are exacerbated as the file gets larger. A more
appropriate data listing would appear in decimal time starting at zero time. Such a data form
facilitates data analysis by conventional numerical integration, such as is done using the computer
-------
program, QTRACER2 (Field, 2002), or by fitting models to the data, such as with the model
CXTFIT2 (Toride et al., 1995).
1.2.2. When to Start Counting
Decimal time should start at zero time, where zero time is most appropriately equated with the
date and time of injection. All subsequent dates and times should then proceed sequentially. Any
odd data interrupts, such as block headers and large interrupts in dates and/or times, should not
prevent proper data counting and processing.
1.2.3. Data Output
The converted data should be rewritten to a data file that can be easily accessed by such common
ASCII-display programs as Notepad® and Wordpad®. The file can then be copied to data input
files used by other programs, such as QTRACER2 and CXTFIT2.
1.2.4. Data Plotting
Graphical plotting of the data allows trends in the data to be easily observed. This is very
valuable for guiding later aspects of any particular study. The ability to save the graphical display
in common file types and in vectorial and raster formats facilitates dissemination to interested
parties.
-------
2. MAIN PROGRAM DESIGN
The new program, FLOWTHRU, was developed with the intention of meeting all the previously
suggested improvements. The program was written in standard FORTRAN, and the executable
form can be run on any PC using Windows. It is fast, simple, and straightforward to run.
Numerous subroutines included with the main program facilitate processing speed and program
clarity.
2.1. MAIN PROGRAM
The main program reads all the necessary input data, processes the measured time values, and
produces a file of decimal time values with corresponding concentration values suitable for use
in breakthrough curve (ETC) analysis. The flowchart depicted in Figure 2 illustrates the basic
flow FLOWTHRU, which is described below. For any new program such as FLOWTHRU to be a
substantial improvement over other methods, it should be able to work almost seamlessly with
the Model 10-AU-005 logged data file downloaded to a computer using the !DL_lBl program.
For example, when FLOWTHRU reads through the Model 10-AU-005 header information, it
acknowledges whether time units were recorded in seconds (sec) or minutes (min), the time
spacing for the recorded units, and the concentration units selected, which can be any one of the
28 different concentration units shown in Table 2. It will be noted from Table 2 that, whereas the
Model 10-AU-005 logged data file records everything in ASCII format suitable for downloading,
FLOWTHRU rewrites these into a slightly more correct scientific form.
2.1.1. Background Concentration
FLOWTHRU accepts the date and time of tracer injection, which is taken as zero time (?j) by
FLOWTHRU. All recorded time data prior to the time of injection is then accorded background
status and is recorded directly to a background-data file with corresponding measured concentra-
tions. Average background concentration Cb prior to tracer injection is added to the background
data file and is obtained from
_ ,
cb = -
Using the data shown in Figure 1 on page 3, FLOWTHRU produced the background data file
shown in Figure 3 on page 10. Figure 3 is fairly self-explanatory in that it shows the date and time
that a dye concentration was measured. At the bottom of the table depicted in Figure 3, the average
background concentration is shown. This latter value is useful because the average background
-------
' Read Process /
Control /—*•
File Name /
Open Process
Control File
Set Interact.
Data File
Processing
/ Read Turner
— »y Designs Data
/ File Name /
Figure 2. Flowchart depicting processing of the program FLOWTHRU.
-------
/ Read Date /
y and Time /-»
/ of Injection /
Nchk = 0
Call Dtrwnd
Call Plot
Screen Only
^o
Figure 2. Flowchart depicting processing of the program FLOWTHRU (cont.).
-------
Table 2. Allowable Turner Designs Model 10-AU-005 concentration units.
Model 10-AU-005 Form
NONE
FSUb
QFTb
PPM
PPB
PPT
mg/1
mg/dl
g/1
nig/kg
mg/g
ug/1
ug/ml
ug/kg
ug/g
ng/1
ng/ml
ng/ul
ng/kg
ng/g
ng/mg
Pg/g
pg/mg
Pg/ug
pg/ml
pg/ul
fg/ul
fg/mg
RAWC
FLOWTHRU Form
Nonea
fsu
qft
ppm
ppb
ppt
mgLT1
mg dLT1
gL"1
mg kg"1
rngg"1
M'gL"1
fJLgmL"1
11 P kg"1
(X& K-5
n-gg"1
ng LT1
ng mL"1
ng (jiLT1
ng kg"1
ngg"1
ng ing"1
Pgg"1
pg nig"1
Pg M-g"1
pgrnL"1
Pg ^L-I
fg jxL"1
fg mg"1
Raw
a A FLOWTHRU plot will not include a display of any concentration units when the Model
10-AU-005 logged data file reads None, but the printed data tables will show "None."
b The Model 10-AU-005 User's Manual does not explain what these units represent. FSU
represents "full scale units" which is more applicable to the old panel meter style unit;
QFT represents the "quantitative fluorescence technique" that was developed by Texaco
for measuring oil fluorescence in exploratory mud samples (J. McCormick, pers. comm.)
c "RAW" is not listed as one of the operational parameters in the Model 10-AU-005 User's
Manual (Turner, 1993, p. A5-3), but it may be selected for calibration purposes (Turner,
1993, p. 3-10) and represents relative fluorescence. FLOWTHRU is set up to read this
"unit" if it can be set as an operational parameter.
-------
BACKGROUND CONCENTRATION
DATE
10/11/02
10/11/02
10/11/02
10/11/02
10/11/02
10/11/02
10/11/02
TIME
(h) (min)
10 24
10
10
10
11
11
11
34
44
54
04
14
24
CONCENTRATION
(M-g/L)
0
0
0
0
0
0
0
.000
.000
.000
.000
.000
.000
.000
AVERAGE BACKGROUND CONCENTRATION = 0.0000
Figure 3. Example background data file produced by FLOWTHRU.
concentration can and should be subtracted from all measured concentrations occurring after dye
injection to better represent actual sample measurements.
If the Turner Designs Model 10-AU-005 data file contains >5 background measurements and
the average background concentration exceeds zero, then a more detailed listing of univariate
statistical parameters is included at the bottom of Figure 3. The main value of this additional
information (not shown in Figure 3 because the background concentration Q,. did not vary) is
to provide the user with a sense of the range of natural background fluorescence. The univariate
statistics, developed using the equations shown in the box in Figure 4, provide a sense of the range
and distribution of values potentially produced when a large number of background measurements
over a relatively long period of time have been taken. Substitution of, for example, the minimum
or maximum background concentration (e.g., Stat(6, *) and Stat(7, *), respectively in Figure 4)
for the average background concentration may be a desirable change in some instances. A generic
form of the univariate statistical output for a varying background concentration (Figure 5) was
developed using data listed in the file, Break.prn (see Section 5.3.2. on page 74 for a brief
discussion of Break.prn). Figure 5 illustrates the univariate statistics that would be created using
the equations shown in Figure 4 .
2.1.2. Measured Water Temperature
If water temperature was measured by a Turner Designs Model 10-AU-005 (see the last two data
columns in Figure 1 on page 3) then FLOWTHRU will average the water temperature data as shown
in Figure 6 on page 12. If water temperature was not measured, then a water-temperature table
is still produced at the end of the background-concentration file (same file as the background-
concentration table) but no water-temperature data are recorded.
10
-------
T2 _ E/i v>i (Xj-
'w ~~ n-\
_
Xw
_ 3
1. Mean, Stat(l, *)
2. Variance, Stat(2, *)
3. Standard Deviation, Stat(3, *)
4. Skewness, Stat(4, *)
5. Kurtosis, Stat(5, *)
6. Minimum, Stat(6, *)
7. Maximum, Stat(7, *)
8. Range, Stat(8, *)
9. Coefficient of Variation, Stat(9, *)
10. Number of values processed, Stat(10, *)
where the statistics are given in terms of a single variable x. The /-th datum is xf, with
corresponding frequencies set to unity (/• = !) and weights also set to unity (w{ = l)a.
= min(xi)
*max =
xcv = *
for x
"Note that none of the variables shown in this box are intended in any way to match similar variables used
elsewhere in this document. For example, *,- here represents any required input parameter (e.g., concentration,
Q,) and « represents the number of measured values used to develop the mean values for each of the required
parameters.
Figure 4. Equations used to calculate the univariate statistics for background concentration
measurements.
Univariate Statistics from UVSTA
Mean
0.0003235
Minimum
0.0000
Variance
0.2210E-06
Maximum
0.001000
Std. Dev.
0.0004701
Range
0.001000
Skewness
0.7544
Coef. Var.
1.4531
Kurtosis
-1.4308
Count
102.0000
Lower CLM Upper CLM Lower CLV Upper CLV
0.0002312
0.0004159 0.1708E-06 0.2973E-06
Figure 5. Univariate statistics created for a Turner Designs data file with varying background
concentrations. Note that CLM and CLV represent the confidence levels for the mean and the
variance, respectively. All other table headings should be readily apparent.
11
-------
AVERAGE MEASURED WATER TEMPERATURE
BACKGROUND WATER SINCE INJECTION ALL MEASURED WATER
14.0 57.2 13.5 56.2 13.5 56.2
Figure 6. Example water temperature data file produced by FLOWTHRU.
Like Figure 3, Figure 6 is fairly self-explanatory in that it shows the average background water
temperature in column 1, the average water temperature from the time of dye injection in column
2, and the average water temperature for all water samples in column 3. Obviously, column 1 will
be the least representative of average water temperature because it is based on the fewest water
sample readings. Water temperature readings are shown in both degrees Celsius (°C) and degrees
Fahrenheit (°F).
2.1.3. Decimal Time Units
The internal data logger in a Model 10-AU-005 limits data intervals to time units of seconds and
minutes, which may not be the most desirable output form for very long tracer experiments (e.g.,
those lasting many days). FLOWTHRU allows the user to set the desired time units to decimal days,
hours, minutes, or seconds (based on a 24-hour clock) regardless of what was set in the internal
data logger.
This operation greatly simplifies an evaluation of a tracer test because inordinately long tracer
tests may be examined with more convenient time values. Tracer experiments <5 hours are
probably best evaluated using the time setting used to log the data (e.g., seconds or minutes).
However, tracer experiments >5 hours may warrant adjusting the time setting to hours or days for
a more readable output.
2.1.3.1. Decimal Time Calculation. Calculation of decimal time is a straightforward rework-
ing of conventional date and time data. After subtracting out background data, FLOWTHRU reads
the first date and time values that correspond with injection date and time. Using a modification
of the method developed by Mull et al. (1988), the date and time values are then converted to a
"start" time tst (time of injection) by
12
-------
so that actual decimal time values are developed according to
where d is incremented by one for every instance that th equals 24 hours (which is reset to zero
in the Turner Designs data file) in the time-concentration file and tu adjusts t for the desired time
units according to Table 3. An examination of Equations (2) and (3) shows that t\ = tst = 0 for
the first measured time value at or immediately after the date and time of tracer release.
Table 3. Multiplier for decimal time units.
Time Units tu
Days j?
Hours 1
Minutes 60
Seconds 3600
2.1.4. Data Counting
To facilitate data processing while minimizing overhead in FLOWTHRU, the data are counted from
the date and time of injection so that the necessary storage space may be dynamically allocated.
The storage space is dependent on the form that the user chooses for the converted data. Processing
ALL the data, a PERCENTAGE of the data, an EXACT-time spacing of the data, or an EVEN-time
spacing of the data are options. If the user selects a processing option other than ALL, then the user
may select to have the data averaged2 and a standard deviation calculated for the concentration
values, which will be displayed as error bars on the final plot, or the user may elect to have the
concentration data "smoothed" using a moving average filter. Alternatively, the user may elect to
just have FLOWTHRU skip to matching time-concentration data.
If just a percentage of the data is selected, then the user is given the option of having
FLOWTHRU select the optimum data percentage to display. Alternatively, the user may manually
select the desired data percentage, in whole numbers only, to display.
2In general, data averaging may not be considered greatly desirable because it is actually taking an average of an
average. In fact, the Turner Designs Model 10-AU-005 can be specifically instructed to average concentration data
for each recorded time measurement.
13
-------
Selection of exact-time or even-time spacing is allowed for daily and hourly decimal times
only. An exact-time spacing is an exact listing of time readings, but data interrupts may cause
uneven spacing.
Even-time spacing is a perfect spacing of time readings, but concentration values may not
perfectly align with the listed times. Inconsistent listing of even-spaced times with concentration
values will generally be insignificant, especially with long-term experiments and/or only minor
time interruptions. This setting, however, may cause adverse effects for short-term experiments or
when large time interruptions occur because insufficient data would be produced for processing.
Even-time spacing should not be used when applying inverse models to the data.
2.1.5. Time Data Processing
After memory requirements have been established and the time to start data processing has been
determined, FLOWTHRU enters a loop mode. It then reads in the selected time data, bypasses all
time interrupts while maintaining proper time spacing, converts the time data to decimal time
in the chosen units, and writes the decimal time data to a user-specified output file. If the
time-concentration data are to be displayed directly on the computer monitor, FLOWTHRU then
automatically calls either NOTEPAD® or WORDPAD®, depending on file size. (More modern PCs
and operating systems are capable of displaying a large amount data on screen using NOTEPAD®
only.) After processing the data, FLOWTHRU plots the data, offers the user the opportunity to have
the background concentration/water temperature tables file and the decimal time-concentration
data file automatically printed, and then closes out the program.
2.1.6. Data Plotting
Upon completion of time data conversion to decimal time, FLOWTHRU calls a data plotting
program and correctly plots the data with the axes appropriately labeled. This file can be manually
saved as a color bitmapped file (rater format) or as a Postscript file (vector format) in either black-
and-white or color formats. Color PostScript files can have either a light blue background or a
black background.
2.1.6.1. PostScript File Descriptions. Three Postscript file formats are provided in FLOWTHRU
in order to meet selected presentation demands by the user. For typical document presentation, a
simple black-and-white display may be desirable in most instances because there is no competing
information (e.g., multiple data displays) on the plot that could cause confusion, memory require-
14
-------
ments are smaller, and there is no need to access a color printer. In other circumstances, a color
PostScript plot may be desirable.
In some instances, a presentation may benefit from a color plot. The user may select a color
plot with a light blue background, black labels, purple data points, and a red line connecting
the data points. Alternatively, in some instances the user may prefer to display a color plot with a
black background (e.g., a slide presentation), which will have yellow labels, cyan data points, a red
line connecting the data points, and a purple key indicating the number of data values processed.
Figure 7 on the next page shows the three different formats developed by FLOWTHRU using the
Turner Designs data referenced in Table 1 on page 3. In Figure 7 the black background color
plot appears to be the least clear of the three, but this is a consequence of producing the plot at
a reduced size and displaying it in a printed document. The main purpose of producing a color
PostScript plot with a black background would be for slide presentations, in which case the plot
will be much clearer.
As shown in Figure 7, each plot has a key, shown in the upper right-hand corner, that indicates
the number of data points and the percentage of the total number of data points processed. The
plots indicate that 143 data values were processed (Data = 143), which has little meaning to
anyone unfamiliar with the true data size (19,995 data points); the indication that only 0.72%
(Perc = 0 . 7 2%) of the data are displayed is a pretty clear indication that only a small fraction
of the data were processed for these figures. Such a small percentage was chosen in this instance
because it was desirable that the display not be overwhelmed by an excessive amount of data.
In addition, a computer-determined PERCENT of the data (see Section 2.1.4.) was chosen for the
desired data-processing mode, which caused FLOWTHRU to automatically determine that only 143
data values would be processed in this instance.
2.1.7. Memory Usage
For most of the program operation, converted data are written directly to an output file. In those
instances in which the user desires to examine the data directly on screen using either NOTEPAD®
or WORDPAD®, the data are written to a temporary file that is deleted after closing NOTEPAD® or
WORDPAD®. Array storage is necessary, however, for passing the converted data to the plotting
routine, so only two or three storage arrays are required.
In all instances, two arrays are always required: one for decimal time and one for measured
tracer concentration. For those instances in which averaged or smoothed concentration values
are requested, then a third array is required for the display of error bars representing the stan-
dard deviation for each average concentration (confidence limits around smoothed concentra-
15
-------
50 100
Time from injection (d)
50 100
Time from injection (d)
Figure 7. Three PostScript formats produced by FLOWTHRU using the Turner Designs data
referenced in Table 1.
16
-------
tions). Standard-deviation values are shown when on-screen display using either NOTEPAD® or
WORDPAD® is requested but are not written to the output file. Standard-deviation values and/or
confidence limits are not written to the output file because such an additional column of data
would not conform with typical file input requirements of breakthrough-curve analysis programs.
2.1.7.1. Dynamic Memory Allocation. To further minimize excess memory usage, FLOWTHRU
carefully counts through the Turner Designs data file from the time of tracer injection t\ to cal-
culate the maximum amount of space required. The memory size is then further optimized by
adjusting the maximum necessary amount downwards as appropriate for a user selection of a re-
duced data set (e.g., percentage, exact-time data spacing, or even-time data spacing) for conversion
(see Section 2.1.4. on page 13).
In addition, when a reduced data set of either exact-time or even-time data spacing is requested,
FLOWTHRU will recalculate the required memory size according to time, relative to whether the
original data were measured in units of seconds or minutes. Although it may be argued that this
additional recalculation is unnecessary, inclusion of this option facilitates PC multitasking, which
allows a user to utilize other related programs while processing the data.
17
-------
3. PROGRAM SUBROUTINES
To improve program readability and decrease processing time, several subroutines have been
included in FLOWTHRU. The subroutines also effectively reduced the number of programming
lines for repeated operations (e.g., DTRWND; see Section 3.2.1. on page 32). Of the subroutines
included in FLOWTHRU, only the process control subroutine, FiLE_PROC, is described in detail
because it allows for considerable interaction by the program user. All other subroutines are
described only briefly.
3.1. INTERPROCESS CONTROL SUBROUTINE
The original development of FLOWTHRU required that the user respond interactively to queries
presented by FLOWTHRU. Although this option still exists in the final version, an easier option
is now provided in the form of an interprocess control file, which essentially contains all the
necessary control statements required by FLOWTHRU so that the program runs quickly without
any additional user input. The form of the subroutine used to read and process this file is depicted
in Figure 8.
The format of an interprocess control file is fairly short and easy to understand. A comment
line precedes each required data line so that the user has a reminder of what the data line should
contain. Figure 9 illustrates the general form of a typical flowing stream input file, the components
of which are briefly described in Table 4. Note that the Line numbers listed in Figure 9 correspond
to the Line numbers listed in Table 4. No column numbers are provided because free format input
is permitted (i.e., the placement of input items is mostly irrelevant). It is apparent that all odd-
numbered lines listed in Figure 9 and Table 4 are just comment lines and can contain anything
that the user decides should appear on such a line or they can be left "blank." However, in no
instance can one of these lines be removed from the process control file. In addition, Line 2 is
also a comment line, as are any lines that appear after Line 28.
It should also be apparent from Figure 9 that the odd-numbered lines running from Line
9 to Line 27 list the options that may be recorded on the following line. Such included
information helps prevent incorrect entries on the line following the listed comment line. However,
FLOWTHRU specifically checks for incorrect entries, attempts to correct these entries, and then
continues processing the data3.
3Note that all 28 lines must appear in the interprocess control file or program operation will be terminated (this
is the only instance where FLOWTHRU is programmed to terminate because of incorrect file entry). However, any
number of lines greater than 28 may appear in the process control file, but Line 29 and subsequent lines will all be
treated as comments to be ignored by FLOWTHRU.
18
-------
Read New
File Name
Read New
Background
File Name
I Return J
Figure 8. Flowchart depicting subroutine for implementing interprocess control file and error
reporting.
19
-------
Generic Interprocess- Control Input File
1 DATA FILE FOR CONTROLLING DEFAULT INPUT FILE
3 FILE NAME FOR TURNER DESIGNS CREATED DATA FILE
4 Turner Designs Data File
5 FILE NAME FOR FILE OF CONVERTED TIME-CONCENTRATION DATA
6 Decimal time-concentration Data File
1 FILE NAME OF BACKGROUND DATA AND AVERAGE TEMPERATURES
8 Background Data File
9 PREFERRED TIME UNITS FOR DATA PROCESSING ([D]ay, [H]our, [M]inutes, [S]econds [A]11)
10 Day
11 CONTROL FOR DETERMINING IF DATA ARE TO BE DISPLAYED ON SCREEN <[Y]es/[N]o)
12 Yes
13 DESIRED FORM OF THE DATA TO BE PROCESSED ([A] 11, [Plercent, [E]xact, e [V] en)
14 All
15 NUMBER OF DAYS OR HOURS FOR EXACT OR EVEN SPACING (K=Day<=10, 24<=Hour<=240)
16 1
17 CONTROL FOR COMPUTER SELECTED OR USER SELECTED PERCENT DETERMINATION ([C]omputer/[U]ser)
18 Computer
19 READ USER-SELECTED PERCENTAGE AS APPROPRIATE (1 <= % <= 100)
20 10
21 CONTROL FOR MATCHING, SMOOTHING, OR AVERAGING DATA ([M]atch/[S]mooth/[A]verage)
22 Match
23 DATE AND TIME OF TRACER INJECTION (MM/DD/YY HH MN)
24 10/11/02 1025
25 POSTSCRIPT FILE CREATION CONTROL ([Y]es/[N]o)
26 Yes
27 FORM OF POSTSCRIPT FILE TO BE CREATED (0, 1, 2, 3)
28 2
30 STOP PROCESSING
31 END OF RUNS
Figure 9. Generic example of interprocess control input file illustrating the basic format used by
FLOWTHRU for processing. Note that the Line numbers are not part of a typical input file and are
listed here only for reference purposes for Table 4.
20
-------
Table 4. Description of the input file components listed in Figure 9.
Line Type Identifier
1
2
3
4 Character FILE NAME
Description
Comment line.
Comment line.
Comment line.
Turner Designs data file.
This is just the actual
5
6
7
8
9
10
11
12
13
14
Character FILE NAME
Character
name of the data file that is downloaded from a
Model 10-AU-005 using the program, !DL_lBl.
Comment line.
Converted time-concentration file name. This is a user-
chosen file name where the newly created decimal time-
concentration data are to be recorded.
Comment line.
Background data file name. This is a user-chosen file
name for recording the background time-concentration
data and the measured temperature data.
Comment line.
TIME UNITS Desired time units to be used when calculating decimal
time ([D]ay, [H]our, [M]inutes, [S]econds, [A]ll).
Character FILE NAME
Character SWITCH
Character FORM
15
Comment line.
Control switch that allows the user to view the
converted (decimal) time-concentration data on the
computer monitor if desired.
Comment line.
Form of the data for the converted (decimal) time-
concentration data file. (Exact and even options
ignored if Line 10 equals [M] inutes, [S] econds,
or [A] 11.)
Comment line.
continued on next page
21
-------
Table 4. Description of the input file components (continued).
Line Type Identifier
Description
16
17
18
19
20
21
22
23
24
25
26
27
28
Integer SPACING
Character SWITCH
Integer PERC.
Character SWITCH
Character SWITCH
Integer
User-selected number of days (1 < As/rf < 10) or hours
(24 < Astfl < 240) to be used when exact- or even-time
spacing is desired. (Ignored if Line 10 equals [M] inutes,
[S]econds, or [A] 11.)
Comment line.
Control switch for allowing FLOWTHRU to determine the
"optimal" percentage of data to process or to allow the user
to select a desired percentage to process.
Comment line.
User-selected percentage of data to process — only whole
numbers allowed. (Ignored if Line 18 equals Computer or
Line 14 does not equal Percent.)
Comment line.
Control switch for selecting measured concentration data,
smoothed concentrations, or averaged concentrations.
(In most instances the user will want to select, [M] atch;
less often, [S] mooth; and rarely [A] verage.)
Comment line.
Date and time of tracer injection in the form of
MM/DD/YY HH MN SS, or MM DD YY HHMN S3.
(Note that time may appear as either HH MN SS or HHMN SS.)
Comment line.
Control switch that allows the user to have a PostScript
plot file created if desired.
Comment line.
Pos tScript User-choice for determining form of PostScript file to be
created. (Ignored if Line 26 equals [N] o)
Character INJECTION
continued on next page
22
-------
Table 4. Description of the input file components (continued).
Line Type Identifier
29
30
31
Description
Comment line.
Comment line.
Comment line.
3.1.1. Line-by-Line Description of Input Files
Figure 9 and Table 4 include a Line number to identify the line on which a particular item must
be supplied in a data input file. These Line numbers are provided here only as a guide to the user
and are never to be included in any actual data input file. Below is a detailed description of each
line for a typical data input file.
Lines 1 and 2 are comment statements provided for the user to enter clarifying statements that
are ignored by FLOWTHRU. These lines are generally used as a convenient identifier and
input file line break, both of which have no meaning.
Line 3 is a comment statement provided for the user to enter a clarifying statement that is ignored
by FLOWTHRU. It is generally used as a convenient label for Line 4, which is READ by the
program. For example, Figure 9 lists Line 3 as:
FILE NAME FOR TURNER DESIGNS CREATED DATA FILE
which is stating that Line 4 is a file name to be READ by FLOWTHRU. Line 3 can be left
blank if desired.
Line 4 is a required user-supplied file name (e.g., Turner Designs Data File) to be READ by
FLOWTHRU. It is intended to be an identifier for the specific tracer test data being collected.
For example, Figure 9 lists Line 4 as:
Turner Designs Data File
23
-------
which should match all references to the actual data file. A default name is given as:
Fluor_time. prn.
Line 5 is a comment statement provided for the user to enter a clarifying statement that is ignored
by FLOWTHRU. It is generally used as a convenient label for Line 6, which is READ by the
program. For example, Figure 9 lists Line 5 as:
FILE NAME FOR FILE OF CONVERTED TIME-CONCENTRATION DATA
which is stating that Line 6 is a file name to be READ by FLOWTHRU. Line 5 can be left
blank if desired.
Line 6 is a required user-supplied file name to be READ by FLOWTHRU. It is intended to be the
name of the file that all the newly created decimal time-concentration data get written to by
FLOWTHRU. For example, Figure 9 lists Line 6 as:
Decimal time-concentration Data File
A default name is given as: Fluor_time. cnv.
Line 7 is a comment statement provided to the user to enter a clarifying statement that is ignored
by FLOWTHRU. It is generally used as a convenient label for Line 8, which is READ by the
program. For example, Figure 9 lists Line 7 as:
FILE NAME OF BACKGROUND DATA AND AVERAGE TEMPERATURES
which is stating that Line 8 is a file name to be READ by FLOWTHRU. Line 7 can be left
blank if desired.
Line 8 is a required user-supplied file name to be READ by FLOWTHRU. It is intended to be the
name of the file that all the background time-concentration data and temperature data get
written to by FLOWTHRU. For example, Figure 9 lists Line 8 as:
Background Data File
A default name is given as: Fluor_time .bak.
Line 9 is a comment statement provided for the user to enter a clarifying statement that is ignored
by FLOWTHRU. It is generally used as a convenient label for Line 10, which is READ by the
24
-------
program. For example, Figure 9 lists Line 9 as:
PREFERRED TIME UNITS FOR DATA PROCESSING ( [D] ay, [H]our, Minutes, [Sleconds, [A] 11)
which is stating that Line 10 is a data item to be READ by FLOWTHRU. Line 9 can be left
blank if desired.
Line 10 is a required data entry item to be READ by FLOWTHRU. As indicated by Line 9 above
(and in Figure 9 and Table 4), Line 10 can be either [D] ay, [H] our, [M] inutes,
[S] econds, or [A] II4. However, the items to be listed on Line 10 can also appear as D,
H, M, S, or A in upper or lower case letters because only the first letter entered anywhere on
Line 10 is actually READ by FLOWTHRU. For example, Figure 9 lists Line 10 as:
Day
which means that time-data processing by FLOWTHRU should be in decimal days.
Line 11 is a comment statement provided for the user to enter a clarifying statement that is
ignored by FLOWTHRU. It is generally used as a convenient label for Line 12, which is
READ by the program. For example, Figure 9 lists Line 11 as:
CONTROL FOR DETERMINING IF DATA ARE TO BE DISPLAYED ON SCREEN ([Y]es/[N]o)
which is stating that Line 12 is a data item to be READ by FLOWTHRU. Line 11 can be left
blank if desired.
Line 12 is a required data entry item to be READ by FLOWTHRU. As indicated by Line 11 above
(and in Figure 9 and Table 4), Line 12 must appear as either Yes [Y] or No [N] anywhere
on Line 12 (case does not matter). For example, Figure 9 lists Line 12 as:
Yes
which means that, after processing by FLOWTHRU, the new time-concentration data file
4Note [A] 11 here only means that data processing will be in Seconds and the x-axis of the plot will be in days,
hours, minutes, and/or decimal seconds (DD HH MM SS.S) as exponents as appropriate (e.g. 10 days, 2 hours, 35
minutes, and 12.4 seconds could appear as 10d2h35m12.4s). In most instances, just a single time unit (e.g., 10d) will
appear under each major tick on the x-axis.
25
-------
should be printed to the computer monitor using either NOTEPAD® or WORDPAD®.
13 is a comment statement provided for the user to enter a clarifying statement that is
ignored by FLOWTHRU. It is generally used as a convenient label for Line 14, which is
READ by the program. For example, Figure 9 lists Line 13 as:
DESIRED FORM OF THE DATA TO BE PROCESSED ([A] 11, [P]ercent, [E]xact, e [V] en)
which is stating that Line 14 is a data item to be READ by FLOWTHRU. Line 13 can be left
blank if desired.
14 is a required data entry item to be READ by FLOWTHRU. As indicated by Line 13
above (and in Figure 9 and Table 4), Line 14 must appear as All [A] , Percent [P] ,
Exact [E] , or Even [V] anywhere on Line 14 (case does not matter). For example,
Figure 9 lists Line 14 as:
All
which means that all the measured time-concentration data are to be processed by FLOWTHRU.
Line 15 is a comment statement provided for the user to enter a clarifying statement that is
ignored by FLOWTHRU. It is generally used as a convenient label for Line 16, which is
READ by the program. For example, Figure 9 lists Line 15 as:
NUMBER OF DAYS OR HOURS FOR EXACT OR EVEN SPACING (K=Day<=10, 24<=Hour<=240)
which is stating that Line 16 is a data item to be READ by FLOWTHRU. Line 15 can be left
blank if desired.
Line 16 is a required data entry item to be READ by FLOWTHRU. As indicated by Line 15 above
(and in Figure 9 and Table 4), Line 16 must appear as a numerical value ranging from 1 to
10 for spacing in days or as a numerical value ranging from 24 to 240 for spacing in hours
anywhere on Line 165. For example, Figure 9 lists Line 16 as:
Line 16 is ignored when Line 10 does not equal Day [D] or Hour [H]
26
-------
which means that all the measured time-concentration data are to be processed by FLOWTHRU
on a single daily basis.
Line 17 is a comment statement provided for the user to enter a clarifying statement that is
ignored by FLOWTHRU. It is generally used as a convenient label for Line 18, which is
READ by the program. For example, Figure 9 lists Line 17 as:
CONTROL FOR COMPUTER SELECTED OR USER SELECTED PERCENT DETERMINATION ([C]omputer/[U]ser)
which is stating that Line 18 is a data item to be READ by FLOWTHRU. Line 17 can be left
blank if desired.
Line 18 is a required data entry item to be READ by FLOWTHRU. As indicated by Line 17 above
(and in Figure 9 and Table 4), Line 18 must appear as either Computer [C] or User [U]
anywhere on Line 186 (case does not matter). For example, Figure 9 lists Line 18 as:
Computer
which means that if a percentage of the data is to be processed by FLOWTHRU, then
FLOWTHRU is tasked with determining the "optimum" percentage of data to process.
Line 19 is a comment statement provided for the user to enter a clarifying statement that is
ignored by FLOWTHRU. It is generally used as a convenient label for Line 20, which is
READ by the program. For example, Figure 9 lists Line 19 as:
READ USER-SELECTED PERCENTAGE AS APPROPRIATE (1 <= % <= 100)
which is stating that Line 20 is a data item to be READ by FLOWTHRU. Line 19 can be left
blank if desired.
Line 20 is a required data entry item to be READ by FLOWTHRU. As indicated by Line 19 above
(and in Figure 9 and Table 4), Line 20 must appear as a numerical value ranging from 1 to
100 (in whole numbers only) anywhere on Line 207. For example, Figure 9 lists Line 20 as:
6Line 18 is ignored when Line 14 does not equal Percent.
1Line 20 is ignored when Line 14 does not equal Percent [P] and/or Line 18 does not equal User [U].
27
-------
10
which means that only 10% of the measured time-concentration data are to be processed by
FLOWTHRU if allowed by Lines 14 and 18.
Line 21 is a comment statement provided for the user to enter a clarifying statement that is
ignored by FLOWTHRU. It is generally used as a convenient label for Line 22, which is
READ by the program. For example, Figure 9 lists Line 21 as:
CONTROL FOR AVERAGING TIME DATA OR MATCHING TIMES ([M]atch/[S]mooth/[A]verage)
which is stating that Line 22 is a data item to be READ by FLOWTHRU. Line 21 can be left
blank if desired.
Line 22 is a required data entry item to be READ by FLOWTHRU. As indicated by Line 21 above
(and in Figure 9 and Table 4), Line 22 must appear as either Match [M], Smooth [S],
or Average [A] anywhere on Line 22 (case does not matter). The parameter Match [M]
tells FLOWTHRU to use matching measured concentrations, parameter Smooth [S] tells
FLOWTHRU to use a moving average to smooth the measured concentrations, and parameter
Average [A] tells FLOWTHRU to average the measured concentrations. For example,
Figure 9 lists Line 22 as:
Match
which means that, of the time data to be processed by FLOWTHRU, the measured concentra-
tion data are to be used.
Line 23 is a comment statement provided for the user to enter a clarifying statement that is
ignored by FLOWTHRU. It is generally used as a convenient label for Line 24, which is
READ by the program. For example, Figure 9 lists Line 23 as:
DATE AND TIME OF TRACER INJECTION (MM/DD/YY HH MN SS)
which is stating that Line 24 is a data item to be READ by FLOWTHRU. Line 23 can be left
blank if desired.
28
-------
Line 24 is a required data entry item to be READ by FLOWTHRU. As indicated by Line 23
above (and in Figure 9 and Table 4), Line 24 must appear with a date and time of injection
anywhere on Line 24. (NOTE: Seconds, SS, need not be entered.) For example, Figure 9
lists Line 24 as:
10/12/02 1025
which is the date and time of tracer injection.
Line 25 is a comment statement provided for the user to enter a clarifying statement that is
ignored by FLOWTHRU. It is generally used as a convenient label for Line 26, which is
READ by the program. For example, Figure 9 lists Line 25 as:
POSTSCRIPT FILE CREATION CONTROL <[Y]es/[N]o)
which is stating that Line 26 is a data item to be READ by FLOWTHRU. Line 25 can be left
blank if desired.
Line 26 is a required data entry item to be READ by FLOWTHRU. As indicated by Line 25 above
(and in Figure 9 and Table 4), Line 26 must appear as either Yes [Y] or No [N] anywhere
on Line 26 (case does not matter). For example, Figure 9 lists Line 26 as:
Yes
which means that after processing by FLOWTHRU, a PostScript time-concentration data plot
file should be created.
Line 27 is a comment statement provided for the user to enter a clarifying statement that is
ignored by FLOWTHRU. It is generally used as a convenient label for Line 28, which is
READ by the program. For example, Figure 9 lists Line 27 as:
FORM OF POSTSCRIPT FILE TO BE CREATED (0, 1, 2, 3)
which is stating that Line 28 is a data item to be READ by FLOWTHRU. Line 27 can be left
blank if desired.
29
-------
28 is a required data entry item to be READ by FLOWTHRU. As indicated by Line 27 above
(and in Figure 9 and Table 4), Line 28 must appear as either 0,1, 2, or 3 anywhere on Line
288. Option 0 means no PostScript file is created — Line 28 is overridden if Yes [Y] is
listed on Line 28. Option 1 means that a black-and-white PostScript file is to be created if
a Yes [Y] occurs on Line 28. Option 2 means that a color PostScript file with a light blue
background is to be created if a Yes [Y] occurs on Line 28. Option 3 means that a color
PostScript file with a black background is to be created if a Yes occurs on Line 28. For
example, Figure 9 lists Line 26 as:
which means that after processing by FLOWTHRU, a color PostScript time-concentration
data plot file with a light blue background (see Section 2.1.6. and Figure 7) should be cre-
ated.
3.2. ARRAY-SIZE ALLOCATION ADJUSTMENT SUBROUTINE
In order to minimize memory requirements, FLOWTHRU specifically allocates the amount of
memory required. The program determines the necessary array sizes to allocate for those instances
for all the data or for those instances where only a portion of the measured time-concentration data
are to be processed (e.g., Percentage). This allows FLOWTHRU to use less memory, which
is mostly an issue for older computers but is still a concern in some instances. To allocate an
appropriate memory size, FLOWTHRU first determines what time spacing and time units were used
to log the data, as listed in the header information in the Turner Designs data file (see Figure 1 on
page 3). Then depending on whether time was recorded in seconds or minutes and if only a portion
of the data is to be processed, FLOWTHRU calls either subroutine SUNIT or subroutine MUNIT to
adjust memory allocation. Memory size is adjusted down from a full count of all data recorded
according to the time spacing and units used to log the data. The subroutine flow for subroutine
SUNIT is described in Figure 10.
The values N4 and N7 shown in Figure 10 are used by FLOWTHRU to adjust memory
allocation. For time units recorded in minutes, subroutine MUNIT appears similar to SUNIT, but
all values listed for N4 are reduced by a factor of 60. For those instances when decimal time is
to be in units other than days (e.g., hours), the calculated value for N4 is increased by adding the
8Ignored if No [N] appears on Line 26.
30
-------
Figure 10. Flowchart depicting subroutine for memory size determination based on seconds.
31
-------
value for N7.
3.2.1. Data-Counting Control
After determining if memory requirements are to be reduced from the maximum, the measured
time-concentration data are counted while by-passing the initial Turner Designs data file header
information and all data block separators. The header separator blocks are by-passed by use of
subroutine DTRWND the flow of which is described in Figure 11.
Figure 11. Flowchart depicting subroutine for counting past Turner Designs header and separate
data blocks.
3.2.2. Data Time Breaks
After calling DTRWND to adjust data counting, FLOWTHRU compares the listed date and time
values with the date and time of injection so that background may be properly established. If
odd and/or long time breaks occur in the data (as occur when the logger is shut down for data
downloading or when a power outage occurs), then FLOWTHRU calls the subroutine TIMCNT.
Subroutine TIMCNT reads through the time break and apportions the necessary missing time to
the decimal time-concentration data file so that an accurate passage of time may be provided. See
Figure 12 for a flowchart illustrating the process used by TIMCNT.
3.2.2.1. Leap Year Calculation. In some instances, data time breaks may cross into and
beyond the month of February, which, rarely, has 29 calendar days as a result of a leap year.
32
-------
Figure 12. Flowchart depicting subroutine for counting past time breaks in the Turner Designs
data file.
33
-------
-o
Figure 13. Flowchart depicting subroutine for determining whether time breaks in the data
occurred in February during a leap year.
Subroutine LEAP (Figure 13), accessed by subroutine TIMCNT, determines whether the given year
is a leap year. Figure 13 illustrates how subroutine LEAP determines whether the year is a leap
year according to the following rules:
Rule 1. Every year exactly divisible by 4 is a leap year (1996 was a leap year).
Rule 2. Except those years exactly divisible by 100 are not leap years (1900 was not a leap year).
Rule 3. Except those years exactly divisible by 400 are leap years (2000 was a leap year).
Subroutine LEAP properly implements all of these rules correctly. Many current spreadsheet
programs follow only rules 1 and 2, but not 3, so they treat the year 2000 as a "common year"
rather than as a leap year. (The existence of "corrective macros" have actually prevented the
underlying bug from being fixed.) Many other programs use only rule 1, so that they incorrectly
consider the year 1900 to be a leap year, even though they get the year 2000 right, by accident
(ACCC, 1998).
Subroutine LEAP was implemented in FLOWTHRU by utilizing the FORTRAN requirement of
conducting operations on integers by integers (whole numbers). For example, when the year 2000
is divided by 4, the result is 500 (in whole numbers), which when multiplied by 4 returns 2000.
The actual determination that the year 2000 is a leap year by subroutine LEAP is as follows:
34
-------
1. 2000/4 = 500
500 x 4 = 2000 [2000 = 2000 -»True] Go to 2
2. 2000/100 = 20
20 x 100 = 2000 [2000 = 2000 -+True] Go to 3
3. 2000/400 = 5
5x400 = 2000 [2000 = 2000 ->True]V Year 2000 is a leap year!
However, when the year 2006 is divided by 4, the result is 501 (in whole numbers), which
when multiplied by 4 returns 2004 rather than 2006. The actual determination that the year 2006
is not a leap year by subroutine LEAP is as follows:
1. 2006/4 = 501
501x4 = 2004 [2004 ^ 2006 -^False]V Year 2006 is not a leap year!
Although subroutine LEAP should be implemented only rarely, it is appropriate that it be done
correctly. The need to accurately calculate solute travel times cannot be overstated.
3.3. DECIMAL TIME FILE AND PLOT FILE HEADERS
To ensure that the results of FLOWTHRU are properly labeled so that confusion regarding time
and concentration units are avoided, FLOWTHRU examines the Turner Designs data file header
for the units originally set when the logger was switched on. Although there are only two
different time units that FLOWTHRU needs to consider (minutes and seconds are the only time
units allowed by the Turner Designs Model 10-AU-005 data logger), there are 28 concentration
units that FLOWTHRU needs to recognize.
Subroutine BUNIT simply takes the listed concentration units and stores the relevant compo-
nents in appropriate variables, which are then used in the plotting routine for y-axis labeling.
Subroutine CUNIT performs the same basic function as BUNIT except that it instead writes a con-
centration header to the converted decimal time-concentration data file.
3.4. TIME-CONCENTRATION DATA AVERAGING
As mentioned in Section 2.1.4. on page 13, time-concentration data averaging should generally
not be done. However, if the user would like to have a set of concentration values averaged up to
a specific time, then the subroutine AVECNT will add the required number of concentration values
and their squared values. Then subroutine AVESTN will solve for the mean concentration and
35
-------
standard deviation, the values of which are recorded as vectors using subroutine ARRAYS. Lastly,
subroutine RESET is called to reset all variables back to zero and to begin the process over.
3.5. MEAN TEMPERATURE CALCULATION
As described in Section 2.1.2. on page 10, FLOWTHRU produces a listing of the mean water
temperature recorded during the tracer test. The mean water temperatures are appropriately
calculated after data processing by FLOWTHRU has been completed and the information passed
to subroutine TEMPER. The mean temperature values are then converted to either degrees Celsius
(°C) or degrees Fahrenheit (°F) depending on the units originally set by the user.
36
-------
4. USING FLOWTHRU TO EVALUATE MODEL 10-AU-005 DATA
FLOWTHRU is an easy-to-use package that is intended to provide the user with a quick
conversion of the Model 10-AU-005 logged data to decimal time with a plot of the BTC so that
trends in the data may be observed. Data processing may occur by reading in an "interprocess-
control file" (described in Section 3.1.) or interactively. Input errors are corrected automatically
by FLOWTHRU with a screen notice to the user so that processing may proceed uninterrupted.
4.1. FLOWTHRU PROGRAM USAGE AND EXAMPLE DATA FILES
NOTE: This program functions best with a display equal to 1024 x 768 pixels, adequately with a
display equal to 800 x 600 pixels, and not so well with further reduced display settings.
Before running the program, all FLOWTHRU files should be copied to the PC's hard drive.
Although plenty of storage space is available on the CD-ROM disk for the creation of data output
files and graphics files, the possibility of damage to the FLOWTHRU program file from excess use
cannot be ignored.
4.1.1. Loading FLOWTHRU and Example Data Files
1. After booting up the computer, place the CD-ROM disk into the computer's CD-ROM disk
drive.
2. At the computer desk top, place the mouse pointer (arrow) on the "My Computer" icon and
click the Right mouse button (Right Click).
3. Left Click on the word "Explore" in the pop-up menu. Alternatively, just hit the letter "E"
on the keyboard.
4. Place the mouse pointer on the CD-ROM disk drive icon (e.g., D: or E:) and Left Double-
Click.
5. Left Click "Edit" at the top of the Window Screen and Left Click on "Select All" in the
pull-down menu. Alternatively, just hit the letter "A" on the keyboard.
6. Left Click on the "Copy" icon on the "Tool Bar" near the top of the Window Screen (second
row). Alternatively, Left Click on "Edit" at the top of the Window Screen and Left Click on
"Copy" or just hit "C" on the keyboard.
7. Left Click on the preferred hard drive (e.g., C:).
37
-------
8. Left Click on the "Paste" icon on the "Tool Bar" near the top of the Window Screen (second
row). Alternatively, Left Click on "Edit" at the top of the Window Screen and Left Click on
"Paste" or just hit "P" on the keyboard.
A folder named FLOWTHRU will be created on the chosen hard drive and all the appropriate
files copied accordingly to the appropriate file folders9.
4.2. FLOWTHRU EXECUTION
FLOWTHRU is very easy to use. After the appropriate data files are created (which are nearly
self-explanatory), FLOWTHRU, for the most part, requires nothing more than hitting as
requested or manipulating the mouse and clicking with the left mouse button. (See Section 3.1. on
page 18 for a detailed discussion of FLOWTHRU data input files.)
1. In Windows Explorer, Left Double-Click the FLOWTHRU folder and then Left Double-
Click the FLOWTHRU.EXE file to initiate program operation.10
2. At this point, FLOWTHRU will open the program initiation screen and title (Figure 14) and
will prompt the user to press to begin full program interaction. Next, the user
is requested to enter an input file name for the file to be evaluated. One advantage of a
subdirectory on the hard disk is that providing an obscure path for all subfiles is not required;
the program will find them automatically because they are all at the same location as the
executable file. If the data files are in different locations from FLOWTHRU, the correct path
to the * . inp files must be provided. Alternatively, pressing will automatically
cause FLOWTHRU to bypass usage of an interprocess control file (e.g., *.inp or any file name
representing a file of the form depicted in Figure 9) and go directly into the interactive mode.
9FLOWTHRU was not designed for MSDOS®use, which requires that the files be moved according to the following
instructions:
• At the C: \> prompt, type "MKDIR FLOWTHRU" (without the quotes — whenever quotes appear in this section,
type the requested information without the quotes).
• Next, copy the executable and data files stored in the file Flowthnudos on the CD to the hard disk (e.g., if C is the
disk drive: "COPY D:\*.* C: \FLOWTHRU\* . *").
• Repeat the above commands for the other files on the CD.
• Put the CD in a safe location.
10If a command prompt is preferred, then at the C: \> prompt type "CD\FLOWTHRU" without the quotes. The
user will then see a new prompt; C: \FLOWTHRU>. The user may now type "FLOWTHRU" to run the program by just
responding to the requested information, assuming that the user has also copied, as created, the necessary data files.
38
-------
* *
* FLOWTHRU.EXE *
* *
* PROGRAM TO CONVERT TURNER DESIGNS MODEL 10-AU-005 *
* SAMPLE TIME DATA TO DECIMAL TIME DATA FOR USE *
* WITH QTRACER2 AND OTHER MODELING PACKAGES *
* *
* DEVELOPED *
*
*
* BY *
* *
* MALCOLM S. FIELD *
* U.S. ENVIRONMENTAL PROTECTION AGENCY *
PRESS RETURN TO CONTINUE
Figure 14. Initial FLOWTHRU screen title, which appears at program start.
39
-------
(Although the interactive mode can be somewhat more tedious to use, it does provide the
user with the opportunity to run FLOWTHRU if an interprocess control file has not been
created.)
3. Now enter a data output file name to be written or press for the default name
(Fluor_time. cnv) as requested. Be aware that previous output files can be overwritten
if the same name is used for more than one. However, the input file previously entered
cannot be overwritten if it is mistakenly reentered.
4. Enter a background file name to be written or press for NULL, which means that
no plot file is to be written (Figure 14). As with the output file name, duplicate usage of plot
file names will result in the overwriting of previous plot files, but existing input file names
will not be accepted. There is no default plot file name used (the default is no plot file). If
a plot file name is given, then two plot files will be created: a time-concentration data file
and a PostScript plot file. The latter will have an appropriate name assigned (e.g., sampling
station name) and a . EPS extension added.
4.3. COMPUTER GRAPHICS
A high-quality color graphics algorithm, PGPLOT11 (Pearson, 1997), which allows cascading
graphics screens, direct printing, creation of screen files, and more using pull-down menus in the
Windows environment, is included in FLOWTHRU. Publication-quality plots may be generated as
PostScript files from the graphics screen incorporated into the program. Alternatively, a screen
dump using any type of printer is possible, and bitmapped plot files may be created "on the fly"
using the pull-down menus.
4.3.1. Features of the Interactive Graphics Loop
Running FLOWTHRU will start a conventional Windows screen with a series of pull-down menus
(Table 5). Each underlined character in Table 5 indicates that the key plus the underlined
character implements the respective menu item. For example, will initiate the pull-
down menu items underneath the File heading. Of course the mouse pointer can be used to access
the menu items.
It is necessary to point out here that most users will not use the pull-down menus very often.
Most of the more useful graphics functions have been built directly into FLOWTHRU so as to
"PGPLOT may be obtained fromhttp: //www.astro, caltech.edu/~tjp/pgplot/
40
-------
alleviate excess work on the part of the user. However, in some instances, the user may find
particular functions of value.
The items shown in Table 5 work whether the program is currently at the text-only screen
(Graphic 1), where the user responds to queries posed by FLOWTHRU, or whether the program is
currently at the data-plot screen (PGPlot Graphics, #1). However, there is little point in accessing
any of the pull-down items from the text-only screen, whereas in the data-plot screen the user may
find some items of value. For example, the color data-plot screen can be printed as it appears,
saved as it appears, resized to fit the whole screen, and so on.
A brief description of each pull-down item shown in Table 5 is provided in the next six
subsections. Because the items are relatively self-explanatory, they are only described briefly.
4.3.1.1. Interactive Data Point Deletion. FLOWTHRU has also been programmed to allow the
user to selectively remove specific data points. This is not a feature that should be used commonly,
but there may be instances in which the user suspects the displayed data point in anomalous (e.g.,
as may occur due to a power surge). It is not a sophisticated routine, it does not always work as
desired, and it should be used with considerable caution.
To selectively remove a specific data point, the user will place the cursor with cross-hairs over
the data point to be removed. Due to the sensitivity of this routine, the user must be very precise
in cursor placement so that the correct point is directly under the cursor (note that FLOWTHRU
may still interpret cursor placement as a point other than that chosen by the user).
Pressing any appropriate keyboard key (e.g., an asterisk [*]) will mark the data point that
FLOWTHRU plans to delete. The marked data point may or may not be the exact data point
selected by the user. Because FLOWTHRU might mark an incorrect data point, the user is required
to respond positively to a FLOWTHRU request to proceed with the data point deletion. To confirm
deletion, the user must first press "Y", and will delete the select data point and redraw
the graph (note that case does not matter, Y = y and N = n). Pressing "N" and will
not delete the selected data point. To select another data point, the user must move the cursor back
to the graphic window and press the Left Mouse Key to activate the graphic window. The actual
process of data point deletion is as follows:
1. After the plot window appears, the user should enlarge the plot screen by Left Clicking on
the "box" in the upper right-hand corner of the screen.
2. Next the user should place the cursor on the blue border at the top of the plot window, press
and hold the left mouse button and drag the plot slightly to the right and down so as to
41
-------
display the text window, where the user will note some basic instructions.
3. At this point the user places the cursor and cross-hairs over a data point on the plot window
to be deleted and presses any keyboard key except the slash (/). The data point to be deleted
is now marked on the plot window and the data coordinates are shown on the text window
with a request confirm deletion.
4. To confirm deletion or cancel deletion, the user must move the cursor over the text window
and Left Click anywhere on the text window to make it active. The user may then enter a
"Y" to confirm deletion or an "N" to cancel deletion followed by the "Enter" key to accept
the command.
5. Entering Y and pressing Enter brings up a requestor asking for the Enter to be pressed
again, which redraws the plot window with the deletion in effect.
6. Entering N and pressing Enter cancels the deletion process and restarts the cross-hair on
the plot window.
7. To select additional points for deletion, the user must move the cursor back over the plot
window and Left Click anywhere on the plot window to make it active.
8. This process may continue as long as the user wants to continue selecting data points for
deletion.
9. At this point the original decimal time-concentration file and PostScript plot file (if one was
to be created) have not been changed. To make the changes (deleted data points) take effect,
the user must place the cursor anywhere over the plot window and press either the slash (/)
or simultaneously press Ctrl - D or Ctrl - Z.
10. The user will see a message asking the user to press the Enter key. However, the user must
first make the text window active by Left Clicking anywhere on the text window. Pressing
Enter then makes all the changes effective and causes FLOWTHRU to proceed to end of
the program, where the last basic commands are displayed.
It will be noted that the decimal time-concentration file and, if requested, a PostScript plot file
will already have been created prior to the interactive routine having been implemented. Although
many data points may have been selectively deleted, no final effect on any formal output files has
42
-------
been performed at this stage. The user may kill the program at this stage by moving the cursor
over the cross (x) in the upper right corner of the screen and pressing the Left Mouse button.
To have FLOWTHRU formally accept all the deleted data points and create a final output
file and/or PostScript plot reflecting these deletions, the user must enter either a slash (/) or
simultaneously press Ctrl - D or Ctrl - Z. FLOWTHRU will then update all files reflecting all the
deletions and proceed to the end of the program.
As a final note, it is not possible to delete the first data point, which always represents the time
of tracer release (t — 0). To be correct, all tracer tests obviously always begin at the time of tracer
release which must be taken as zero time.
4.4. FLOWTHRU SOURCE
The FORTRAN source for FLOWTHRU is included on the CD. It is a fairly large program and was
split into pieces to allow for easier reading. It is not recommended that users attempt to follow the
logic or modify the program. Questions regarding the program's functionality can be addressed
to the author.
In addition, the graphics routine developed at the California Institute of Technology is
included; however, it is not allowed for use in commercial products.
43
-------
Table 5. Pull-down menu items available in FLOWTHRU.
File
Print...
Save...
Exit Ctrl+C
Edit
Select Text
Select Graphics
Select All
Copy Ctrl+Ins
Paste
View
Size To Fit
Full Screen Alt Enter
State Window
Pause Ctrl+S Cascade
Tile
Arrange Icons
Input
Clear Paste
Status Bar
j_Graphic 1
2PGPlot Graphics, # 1
Help
Contents
Using Help
About
-------
File
Print... A screen dump to the local printer attached to the respective PC.
Save... Save the screen as a bitmapped (*.BMP) file.
Exit Ctrl+C Exit the program.
Edit
Select Text Select text for pasting to the clipboard.
Select Graphics Select graphics for pasting to the clipboard.
Select All Select both text and graphics for pasting to the clipboard.
Copy Ctrl+Ins Copy selected items to the clipboard.
Paste Paste selected items to the screen.
View
Size To Fit Fit the graphics screen to the view surface without scroll bars.
Full Screen Alt+Enter Fit the entire graphics screen to the view surface without the menu items
displayed (a left-mouse click returns to the original screen).
State
Pause Ctrl+S Pause the graphic display.
Resume Ctrl+Q Resume graphic display.
Pause and Resume appear only as alternates of each other, so that only the one that is not currently
functioning is accessible. The one that is currently in operation is not displayed in the pull-down
menu.
45
-------
Windows
Cascade Allows for a cascading view of multiple child windows at one time.
Tile Allows for a tile display of multiple child windows at one time.
Arrange Icons Not currently used in FLOWTHRU.
Input Automatically displays the input screen (Graphic 1) for data input.
Clear Paste Clears an item pasted onto the screen.
Status Bar Displays the current operating mode of the displayed graphics screen in a bar at the bottom
of the screen (when "check marked").
i Graphic I Name of the data input screen ("check marked") if active.
2 PGPlot Graphics, # 1 Identifying name/number of all subsequently opened graphics screens (active
when "check marked").
Help
Contents Listing of available help contents.
Using Help Description on the use of Help.
About Identifies the FLOWTHRU program.
46
-------
5. EXAMPLE RESULTS
The program FLOWTHRU is designed to produce decimal-time concentration data files and
plots quickly and easily. Input by the user can be by the somewhat tedious method of interactively
entering responses to questions posed by FLOWTHRU or by entering an interprocess-control file.
(See Section 3.1. on page 18 for a detailed description of how FLOWTHRU uses an interprocess
control file and Figure 9 on page 20 for an example process control file.) In the following
examples, all production was accomplished using a process control file.
5.1. PROCESSING SIMPLE DATA FILES
A simple Turner Designs data file is a file, for the purposes of this section, that does not exhibit
any interruptions in the file. A typical interruption in a Turner Designs data file would be a result
of separate data blocks that have occurred as a result of a scheduled fluorescence data logging.
Two simple Turner Designs data files are included here.
5.1.1. Turner Designs Data File TracqOl.prn
The data file TracqOl.prn12, was produced from a column test with the fluorometer logging data
every second. The entire column experiment lasted a little more than 18 minutes. Figure 15 is a
plot of the time-concentration data file using decimal minutes. From Figure 15 a number of things
are apparent. First, all the data lie significantly above zero concentration. Data recording actually
coincided with tracer injection, so no background concentrations were produced. To evaluate
this data set using a program such as QTRACER2, some "representative" value for background
concentration should be developed using the first seven minutes.
The second apparent item is that considerable noise occurs in the data when the logger is set
to record as frequently as every second. Although the data, as is, can be processed using any
number of different modeling packages, it might be desirable to have a smoothing algorithm (e.g.,
Moving Average Filter, Convolution, etc.) applied to the data to remove the excess noise (see
Section 5.1.3.).
The third apparent item is that there are so much data recorded in Figure 15 it is very difficult
to identify any specific features in the plot. This problem can be overcome by having FLOWTHRU
process a percentage of the data, which will also cause a decrease in the amount of noise in the
12Data file, TracqOl.prn was developed by Anne Motelay-Massei and provided by Nicolas Massei of the
Departement of Geologic, Universite de Roun.
47
-------
data. However, processing just a percentage of the data may cause unwanted aliasing, which must
be considered by the user. Figure 16 is a plot of ~20% of the data.
A close visual inspection of Figures 15 and 16 suggests that there is no significant difference
between the two plots. However, a data-analysis or modeling package (e.g., QTRACER2 or
CXTFIT2) is needed to verify that any differences are minor (as was done in this report in
Sections 5.1.3.2. on page 53 and 5.1.3.3. on page 59). Figure 16 could be displayed with even
less data to allow additional clarity.
For example, Figure 7 on page 16 shows just 0.7% of the data provided in Figure 1 on page 313.
Processing just 0.7% of the data was determined by FLOWTHRU because an exact match to every
decimal day was requested.
5.1.2. l\irner Designs Data File Tracq04.prn
The data file Tracq04.prn14 was produced from a column test with the fluorometer logging data
every second. The entire column experiment lasted a little less than 9 minutes. Figure 17 is a plot
of the time-concentration data file using decimal minutes. Like Figure 15, Figure 17 also shows
that all the data lie significantly above zero concentration. Again, data recording coincided with
tracer injection, so no background concentrations were produced. Breakthrough curve evaluation
requires that some "representative" value for background concentration be developed from the
first two minutes.
In addition, considerable noise occurs in the data shown in Figure 17 when the logger is set
to record as frequently as every second, just as occurred in Figure 15. However, Figure 17 also
exhibits extreme noise at the very start of the ETC, which could be "real" or may be leftover
tracer from a previous tracer test that was sorbed onto particles making up the porous media in
the column. In order to better study Figure 17, a reduced data set can be plotted, as was done with
the TracqOl.prn data in Figure 16. Figure 18 is a plot of just 10% of the Tracq04.prn data with
data averaging. The error bars included in Figure 18 provide an indication of the effect of data
averaging, which is extreme at the beginning of the file and not so significant as the "real" data
are processed. It is up to the user to determine whether data averaging is appropriate.
13The full data file Fluor-Time.prn is on the enclosed CD and is a default file used by FLOWTHRU for testing.
14Data file TracqOl.prn was developed by Anne Motelay-Massei and provided by Nicolas Massei of the
Departement of Geologic, Universite de Roun.
48
-------
0.25
I
0.2
o
0.15
Data = 1092
Perc = 100.0 %
10 15
Time from injection (min)
20
Figure 15. Decimal time-concentration data file for the TracqOl.prn data file developed from a
short-term column study (100% of the data shown).
0.25
E
Q.
I 0.2
o
o
0.15
Data = 219
Perc = 20.1 %
10 15
Time from injection (min)
Figure 16. Decimal time-concentration data file for the TracqOl.prn data file developed from a
short-term column study (20.1% of the data shown).
49
-------
0.25
0.2
o
c
o
O
0.15 L
Data = 530
Perc = 100.0 %
4 6
Time from injection (min)
Figure 17. Decimal time-concentration data file for the Tracq04.prn data file developed from a
short-term column study (100% of the data shown).
0.22 -
0.2 -
o
1
0.18
§
o
0.16 (
Data = 53
Perc = 10.0 %
4 6
Time from injection (min)
Figure 18. Decimal time-concentration data file for the Tracq04.prn data file developed from
a short-term column study (10% of the data shown). Note error bars indicating the effect of
averaging.
50
-------
5.1.3. Moving Average Filter Effect on a Data File
In Section 5.1.1., it was suggested that use of a moving average filter (or similar operation) as a
means for smoothing data logged on an extremely frequent basis (e.g., every second) might be
desirable in rare instances. Tracer data smoothing is generally not recommended, but a moving
average filter (Brandt, 1999, pp. 427-440) has been included in the FLOWTHRU package for those
instances where data smoothing may be desirable. Both the TracqOl.prn and Tracq04.prn data sets
exhibit a significant amount of noise as a result of the frequency of data recording (every second),
so a data-smoothing operation might be desirable for these two data sets. The moving average
filter routine included in FLOWTHRU has been applied to the TracqOl.prn and Tracq04.prn data
sets so that the smoothing effect may be observed.
Figure 19 is a plot of smoothed TracqOl.prn decimal time-concentration data. Comparing
Figure 19 with Figure 15 on page 49 suggests that in this particular instance a moving average
filter might be desirable.
Figure 20 is a plot of smoothed Tracq04.prn decimal time-concentration data. The 95%
confidence limits for the smoothed decimal time-concentration data file for the Tracq04.prn data
are also shown in Figure 20 (orange lines in Figure 20). The confidence limits stand out at the
early-time data but cannot be observed for the rest of the data. This suggests that although much
of the smoothed data vary minimally from the real data, the early-time data developed substantial
errors when smoothed. Figure 19 also has the 95% confidence limits plotted for the smoothed
data, but the errors are so minimal the confidence limits cannot be observed.
Comparing Figure 20 with Figure 17 suggests that in this particular instance a moving average
filter might not be quite as desirable because of the early-time "noise" in the Tracq04.prn data
file. The effect of this early-time noise is reflected in the 95% confidence level (orange lines in
Figure 20). Both Figures 17 and 20 indicate that the early-time data in this file represent a special
problem that needs to be considered.
Peak concentration equals 0.212-0.214 ppm (after subtracting ~Cb = 0.160 ppm) for the data
shown in Figure 17 (see Sample Numbers 165-168 in Appendix I on page 81) and 0.212 ppm for
the smoothed data shown in Figure 20 (see Sample Numbers 167-176 in Appendix I on page 81).
This net difference of 0.000-0.002 ppm, although carried through the smoothed data, may not
represent a substantial error when a quantitative analysis of the ETC is performed. The same may
not be said for the TracqOl.prn data set, where peak measured concentration equals 0.246 ppm for
the data shown in Figure 15 and 0.246 ppm for the smoothed data shown in Figure 19 for a net
difference of 0.000 ppm (after subtracting Cb = 0.156 ppm).
51
-------
0.25
0.2
0.15
Data = 1092
Perc = 100.0%
10
15
20
Time from injection (min)
Figure 19. Smoothed decimal time-concentration data file for the TracqOl.prn data file developed
from a short-term column study (100% of the data shown).
0.2
'0.15
Data = 530
Perc = 100.0 %
V
J 0.1
0.05
Time from injection (min)
Figure 20. Smoothed decimal time-concentration data file for the Tracq04.prn data file developed
from a short-term column study (100% of the data shown). Note confidence limits shown in
orange.
52
-------
In addition, Figure 15 shows that the peak time of arrival occurred at 10.53 min after injection,
whereas Figure 19 shows that peak arrival occurred at 10.60 min (a net difference of 0.07 min).
Figure 17 shows that the peak time of arrival occurred at 2.733-2.783 min, whereas Figure 20
shows that peak arrival occurred at 2.867 min (a net difference of 0.134 - 0.084 min). These time
shifts in the smoothed data suggest the possibility that some time-of-travel error (significant or
otherwise) may be evident. These differences in travel times, although small, may be critically
important when modeling solute transport times. Such differences in travel times can have major
consequences and must be regarded very carefully. In these two instances, both smoothed data
files indicate slower transport times for peak concentrations than that which was measured, but
modeling efforts need to be conducted to determine the significance of the slower transport rates.
5.1.3.1. Absolute and Relative Smoothing Errors. The differences observed between the
measured data and the smoothed data for the TracqOl.prn and the Tracq04.prn data sets are
better illustrated in Figures 21 and 22, respectively. The absolute errors between the measured
concentrations and the smoothed concentrations acquired from the moving average are calculated
by
absolute error = true value - approximate value (4)
and the relative errors between the measured concentrations and the smoothed concentrations
acquired from the moving average are calculated by
absolute error (1-^
relative error = : \?)
true value
The absolute and relative errors between the measured ETC and the smoothed ETC for the
TracqOl.prn data set are illustrated in Figures 23 and 24, respectively. The absolute and relative
errors shown in Figures 23 and 24 appear amplified as a result of the scale chosen.
The absolute and relative errors between the measured ETC and the smoothed ETC for the
Tracq04.prn data set are shown in Appendix I on page 81.Figures 25 and 26 are plots of the
absolute and relative errors, respectively, between the measured ETC and the smoothed ETC for
the Tracq04.prn data set.
5.1.3.2. FLOWTHRU Analysis of the Measured and Modified Data Sets. Before the averaged
or smoothed data sets can be accepted, a careful evaluation of the averaging and smoothing effects
should be conducted. To evaluate the averaging and smoothing effects here, the QTRACER2
program (Field, 2002) was used. First, the measured data were evaluated using QTRACER2 and
53
-------
0.25
i
0.2
0.15
5 10 15
Time from injection (min)
20
Figure 21. Comparison of measured and smoothed concentrations TracqOl.prn data file.
0.25
'£ 0.2
o
o
0.15
4 6
Time from injection (min)
Figure 22. Comparison of measured and smoothed concentrations Tracq04.prn data file.
54
-------
Time from injection (min)
Figure 23. Plot of absolute errors between measured and smoothed concentrations for the
TracqOl.prn data file.
Time from injection (min)
Figure 24. Plot of relative errors between measured and smoothed concentrations for the
TracqOl.prn data file.
55
-------
I
4 6
Time from injection (min)
Figure 25. Plot of absolute errors between measured and smoothed concentrations for the
Tracq04.prn data file.
0.4
-0.2 -
4 6
Time from injection (min)
Figure 26. Plot of relative errors between measured and smoothed concentrations for the
Tracq04.prn data file.
56
-------
Table 6. Evaluation of measured and modified concentration parameters for TracqOl.prn.
Data Set Mass Recovered
(»JLg)
Measured
Averaged2
Smoothed
35.099
35.157
35.002
Percent Recovered
(dimen.)
99.997
100.160
99.722
Time to Peak Peak Concentration
(seconds) (PPm)
639.000
640.000
626.000
0.091
0.091
0.090
FLOWTHRU-determined percentage of data to process (= 50%).
Transport distance L = 1.20 m.
Discharge flow Q - 0.006 m3 h~'.
Background concentration Cfc =0.156 ppm (subtracted from data file).
Table 7. Evaluation of measured and modified transport parameters for TracqOl.prn.
Data Set First Arrival3 Mean Travel Time
(seconds) (seconds)
Measured
Averaged"3
Smoothed
0.0000
0.0000
0.0000
667.320
667.440
673.260
Mean Velocity
(ms-1)
0.00180
0.00180
0.00178
Axial Dispersion
(m2 s-1)
2.10 x
1.51 x
2.12 x
10 5
io-5
io-5
Peclet No.
(dimen.)
102.68
143.10
100.91
a Residual tracer in the system incorrectly causes an apparent first arrival equal to injection time.
b FLOWTHRU-determined percentage of data to process (= 50%).
Transport distance L = 1.20 m.
Discharge flow Q = 0.006 m3 h"1.
Background concentration C6 =0.156 ppm (subtracted from data file).
then the averaged and smoothed data were evaluated. The evaluation results for the TracqOl.prn
data set are shown in Tables 6 and 7 and the evaluation results for the Tracq04.prn data set are
shown in Table 8 and 9.
Table 6 shows that the results of both the averaged and the smoothed data sets reasonably
match the measured results even though a slightly greatermass of dye was recovered for the
averaged concentrations. Table 7 also shows acceptable matches between the averaged and
smoothed data sets with the measured data set. Only the mean travel time and axial dispersion
values for the smoothed data set deviate to any significant degree from the measured data set.
Tables 8 and 9 are more interesting than Tables 6 and 7 because the smoothed data sets show
what might be considered better parameter estimates for peak time of arrival tp, axial dispersion
57
-------
Table 8. Evaluation of measured and modified concentration parameters for Tracq04.prn.
Data Set Mass Recovered
(M-g)
Measured
Averaged2
Smoothed
25.343
25.284
24.913
Percent Recovered Time to Peak Peak Concentration
(dimen.) (seconds) (ppm)
100.010
99.779
98.313
9.002
10.002
172.000
0.090
0.079
0.052
FLOWTHRU-determined percentage of data to process (= 50%).
Transport distance L = 1.20 m.
Discharge flow Q = 0.024 m3 h"1.
Background concentration Cj =0.160 ppm (subtracted from data file).
Table 9. Evaluation of measured and modified transport parameters for Tracq04.prn.
Data Set
Measured
Averaged3
Smoothed
First Arrival
(seconds)
6.0017
6.0017
8.0017
Mean Travel Time
(seconds)
152.808
152.664
156.552
Mean Velocity
(ms"1)
0.00785
0.00786
0.00767
Axial Dispersion
(m2 s-1)
1.01 x 10-3
9.77 x 10~4
6.12 x 1(T5
Peclet No.
(dimen.)
9.35
9.65
150.35
FLOWTHRU-determined percentage of data to process (= 50%).
Transport distance L — 1.20 m.
Discharge flow Q = 0.024 m3 h~'.
Background concentration ~Cb =0.160 ppm (subtracted from data file).
Dx, and the Peclet number Pe than did the measured or averaged data set. All three forms of the
Tracq04 data sets resulted in approximately the same mass recoveries, first arrival times, mean
time of travel, and mean velocity, but the similarities end there. The measured time to peak
concentration (tp = 9.000 s) and averaged time to peak concentration (tp - 10.002 s) are clearly
in error, whereas the smoothed time to peak concentration (tp - 172 s = 2.867 min) is much more
reasonable (see Figure 17 on page 50). The incorrect estimate for tp for the measured data by
QTRACER2 is a direct consequence of the extreme noise in the early-time data in Tracq04.prn and
should be apparent when examining a plot of the data (Figure 17 on page 50). This early-time
noise in the data could be manually corrected.
Additionally, the parameter estimates for the TracqOl data sets may be taken as more reliable
than the Tracq04 data sets for the flow system because the TracqOl data sets do not have such
58
-------
extreme early-time noise as the Tracq04 data sets. If this assumption is acceptable, then it
follows that the estimates for axial dispersion (Dx = 6.12 x 1(T5 m2 s"1) and Peclet number
(Pe = 150.35) are more reasonable than are the estimates derived from the Tracq04 measured and
averaged data sets.
5.1.3.3. CXTFIT2 Analysis of the Measured and Modified Data Sets. To further investi-
gate the impact of smoothing data automatically logged by a Turner Designs Model 10-AU-005,
CXTFIT2 (Toride et al., 1995) was run in the inverse mode for both the equilibrium and nonequi-
librium models. The intent was to calibrate the FLOWTHRU-estimated parameters to theoretical
models to see how well the modified data results matched measured data results.
Figures 27, 28, and 29 are plots of the equilibrium model fits to the measured, averaged, and
smoothed TracqOl data sets, respectively. Figures 30, 31, and 32 are plots of the nonequilibrium
model fits to the measured, averaged, and smoothed TracqOl data sets, respectively. All six model
fits appear quite good visually.
Figures 33, 34, and 35 are plots of the equilibrium model fits to the measured, averaged, and
smoothed Tracq04 data sets, respectively. Figures 36, 37, and 38 are plots of the nonequilibrium
model fits to the measured, averaged, and smoothed Tracq04 data sets, respectively. Unlike the
TracqOl.prn data sets (Figures 27-32), only Figures 35 and 38 appear good visually.
Parameter estimates for the equilibrium model fits for the TracqOl.prn and Tracq04.prn data
sets are shown in Tables 10 and 11, respectively. The data shown in Table 10 show relatively little
difference between the measured, averaged, and smoothed methods. However, the smoothed data
shown Table 11 is much different than the measured and averaged data, which is a direct result of
the improved model fitting that occurs with the smoothed data.
Parameter estimates for the nonequilibrium model fits for the TracqOl.prn and Tracq04.prn
data sets are shown in Tables 12 and 13, respectively. Again, the smoothed data shown in Table 13
are very different from the measured and averaged data and represent an improvement.
5.2. PROCESSING MODERATELY COMPLEX DATA FILES
As described in Section 1.1.1. on page 2 the process of downloading a file to a computer from a
Model 10-AU-005 by use of the program iDL_lBl may result in a slight time break in the data
file because of the necessity of shutting down the logger during file downloading. This slight time
break must be properly identified and a correction included in the converted data file.
59
-------
100
0.05
0.1
0.15 0.2 0.25 0.3
Time (h)
Figure 27. Plot of the equilibrium model fit to the measured TracqOl.prn data.
100
Figure 28. Plot of the equilibrium model fit to the averaged TracqOl.prn data.
60
-------
0.05
0.1
0.15 0.2 0.25
Time (h)
Figure 29. Plot of the equilibrium model fit to the smoothed TracqOl.prn data.
100
0.05
0.1
0.15 0.2
Time (h)
0.25
Figure 30. Plot of the nonequilibrium model fit to the measured TracqOl.prn data.
61
-------
100
0.05
0.1
0.15 0.2
Time (h)
0.25
0.3
Figure 31. Plot of the nonequilibrium model fit to the averaged TracqOl.prn data.
0.05
0.25
0.3
0.15 0.2
Time (h)
Figure 32. Plot of the nonequilibrium model fit to the smoothed TracqOl.prn data.
62
-------
80
60
E40
+j
c
0)
o
c
o
O
20
t rv .
0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Time (h)
Figure 33. Plot of the equilibrium model fit to the measured Tracq04.prn data.
80
60
S40
c
o
o
20
L
"0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Time (h)
Figure 34. Plot of the equilibrium model fit to the averaged Tracq04.prn data.
63
-------
50 -
0.02 0.04 0.06 0.08 0.1 0.12 0.14 016
Figure 35. Plot of the equilibrium model fit to the smoothed Tracq04.prn data.
so
60
240
o
o
20
_L
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Time (h)
Figure 36. Plot of the nonequilibrium model fit to the measured Tracq04.prn data.
64
-------
80 -
0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Time (h)
Figure 37. Plot of the nonequilibrium model fit to the averaged Tracq04.prn data.
50
40
J30
S
1
o
°20
10
0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Time (h)
Figure 38. Plot of the nonequilibrium model fit to the smoothed Tracq04.prn data.
65
-------
Table 10. Evaluation of equilibrium model estimated transport parameters for TracqOl.prn.
Data Set Mean Travel Time
(seconds)
Measured
Averaged3
Smoothed
662.983
662.983
666.667
Mean Velocity
(m s"1)
0.00181
0.00181
0.00180
Axial Dispersion
(m2 s"1 )
2.22 x 10~5
2.22 x 1(T5
2.17 x 1(T5
Peclet No.
(dimen.)
97.84
97.84
99.54
R2
0.9977
0.9977
0.9981
FLOWTHRU-determined percentage of data to process (= 50%).
Transport distance L = 1.20 m.
Discharge flow Q = 0.006 m3 rT1.
Background concentration C/, = 0.156 ppm (subtracted from data file).
Table 11. Evaluation of equilibrium model estimated transport parameters for Tracq04.prn.
Data Set
Measured
Averaged3
Smoothed
Mean Travel Time
(seconds)
18.007
20.000
178.306
Mean Velocity
(ms-1)
0.06664
0.06000
0.00673
Axial Dispersion
(m2 s"1)
2.02 x 10~5
1.95 x 10~3
9.71 x 10~5
Peclet No.
(dimen.)
3.96 x 103
3.69 x 101
8.32 x 101
R2
******
******
0.8247
FLOWTHRU-determined percentage of data to process (= 50%).
Transport distance L = 1.20 m.
Discharge flow Q = 0.024 m3 h~'.
Background concentration C/, = 0.160 ppm (subtracted from data file).
66
-------
Table 12. Evaluation of nonequilibrium model estimated transport parameters for TracqOl.prn.
Data Set
Measured
Averaged3
Smoothed
Mean Time
of Travel
(seconds)
666.667
666.667
674.157
Mean
Velocity
(ms^1)
0.00180
0.00180
0.00178
Axial
Dispersion
(m2 s-1)
9.62 x 1(T6
3.67 x 1(T6
3.01 x 10~6
Peclet
Number
(dimen.)
224.53
588.56
709.64
Degree of
Nonequilibrium
(dimen.)
0.8303
0.7118
0.7027
Mass Transfer
Coefficient
(rT1)
4.740
9.564
10.000
R2
0.9989
0.9990
0.9995
a FLOWTHRU-determined percentage of data to process (= 50%).
Transport distance L — 1.20 m.
Discharge flow Q = 0.006 m3 rT1.
Background concentration Cfc =0.156 ppm (subtracted from data file).
Table 13. Evaluation of nonequilibrium model estimated transport parameters for Tracq04.prn.
Data Set
Measured
Averaged3
Smoothed
Mean Time
of Travel
(seconds)
5.072
5.668
343.840
Mean
Velocity
(ms"1)
0.2366
0.2117
0.00349
Axial
Dispersion
(m2 s-1)
1.01 x 10~4
9.77 x 10~5
4.07 x 10~5
Peclet
Number
(dimen.)
2.81 x 103
2.60 x 103
1.03 x 102
Degree of
Nonequilibrium
(dimen.)
0.0930
0.4016
0.5154
Mass Transfer
Coefficient
(h ')
0.2150
0.2064
0.2013
R2
-0.1363
-0.1251
0.8404
FLOWTHRU-determined percentage of data to process (= 50%).
Transport distance L = 1.20 m.
Discharge flow Q = 0.024 m3 h~'.
Background concentration ~Cb =0.160 ppm (subtracted from data file).
-------
5.2.1. 1\irner Designs Data File Creek.prn
The data file Creek.prn was developed from a field study in which the Model 10-AU-005 was set
to log data every 30 min with concentrations reported in ppb. Data downloading was initiated 13
days after tracer injection and a follow-up data downloading approximately one day later. The
first instance of data downloading resulted in a time break of 46 min, 27 s, which translates into
a 16 min, 27 s, logger shutdown time during which logged data were downloaded (actual data
transfer is generally much more rapid than 16 min, but minor difficulties can cause extended time
delays).
Although an ~16.5 min loss of time is relatively insignificant for a tracer test lasting >13
days, it can be significant for tracer tests of a shorter duration. Alternatively, time breaks of a
much greater duration, such as would occur during a power outage, can be very significant for a
tracer test of this duration. Therefore, it is critical that any analysis of the logged data consider
and properly compensate for inopportune time breaks in the logged data file.
Figure 39 is a plot of the time-concentration data file in decimal days using all the data. An
arrow depicting the location of the lost time (~16.5 min) is also shown in Figure 39. Because there
are so much data shown in Figure 39 and because the missing time is so insignificant relative to
the overall duration of the tracer test, it is not visually possible to distinguish where the missing
time exactly occurs.
Figure 40 is a plot of the time-concentration data file in decimal days using < 10% of the data
(FLOWTHRU-determined percentage to process). The arrows shown in Figure 40 now more clearly
indicate the location of missing time relative to the recorded data. From Figure 40 (and less so
from Figure 39) it may be discerned that the break in time recording occurred slightly after 13
days since the time of injection. Also more apparent is the relative insignificance of the missing
data in this particular instance. (Note: The arrows depicted in Figures 39 and 40 are shown here
only for illustration purposes and are not part of the FLOWTHRU package.)
5.3. PROCESSING VERY COMPLEX DATA FILES
Long-term tracer tests can result in some very complex data files because of (1) acquisition of a
large amount of logged data, (2) numerous time breaks in the data reflecting frequent interruptions
in data logging to download data, (3) significant loss of data as a result of a power outage or power
surge causing an interruption in logger operation, or (4) some combination of any of the three
previously identified issues causing complexity.
68
-------
*
Q.
Data = 646
Perc = 100.0 %
Location of Missing
Time (~16.5 min.)
Time since injection (d)
Figure 39. Decimal time-concentration data file for the Creek.prn data file developed from a 2 km
tracer test (100% of the data shown). Note arrow indicating the location of missing time.
2.5
_
a.
a.
o
c
o
O
1.5
0.5
Data = 47
Perc = 7.3
13.5
Location of Missing
Time (~16.5 min.)
I
10
Time since injection (d)
Figure 40. Decimal time-concentration data file for the Creek.prn data file developed from a 2 km
tracer test (<10% of the data shown). Note arrow indicating the location of missing time. Also
note expanded graph inset with missing time more prominently depicted.
69
-------
5.3.1. 1\irner Designs Data File Fluor_time.prn
The data file FluorJime.prn was developed from a field study in which the Model 10-AU-005
was set to log data every 10 min with concentrations reported in |ig Lr1. Data downloading
occurred several times after tracer injection, with numerous time interruptions and 19,995 total
instrument-logged data. Although not all instances of data downloading resulted in significant
logger interruptions during which logged data were downloaded, several of them did result in
fairly long time breaks (e.g., >20 min).
Figure 41 is a plot of the time-concentration data file in decimal days using all the data; the
ETC shown in Figure 41 suggests no time breaks. This is because the very large amount of
data collected over a fairly long period coupled with the fact that the time breaks were limited to
-~20 - 30 min tend to minimize the expected negative effect of missing time-concentration data.
In addition, FLOWTHRU further realistically minimizes the expected negative effect by properly
counting the lost time and adjusting the ETC.
The very large amount of data depicted in Figure 41 makes it very difficult to discern particular
features of the Fluor_time.prn data set. This difficulty may be overcome by instructing FLOWTHRU
to process only a portion of the data. Figure 42 is a plot of the time-concentration data file in
decimal days with FLOWTHRU set for processing the data for each single day (i.e., 0.71% of the
data processed).
Processing may be further enhanced by processing still less data than occur every day. For
example, Figure 43 is a plot of the time-concentration data file in decimal days with FLOWTHRU
set for processing the data every two days (i.e., 0.36% of the data processed). Figure 44 is a plot of
the time-concentration data file in decimal days with FLOWTHRU set for processing the data every
five days (i.e., 0.15% of the data processed), and Figure 45 is a plot of the time-concentration data
file in decimal days with FLOWTHRU set for processing the data every 10 days (i.e., 0.075% of the
data processed). Not readily apparent in Figures 42-45 is that counting past selected time values
at a set frequency will usually result in the last one or more data points not being counted because
it (they) does not match the chosen time spacing As,. FLOWTHRU is designed to recognize this
situation and process the last uncounted data points even though the Ast is not maintained.
Figure 46 is a plot of the time-concentration data file in decimal days with FLOWTHRU set to
process just 2% of the data. Instructing FLOWTHRU of the percent of data to process may be more
desirable in many instances. Figure 46 shows a much more useful amount of processed data (i.e.,
400 data points) while not having to deal with the entire amount of data (i.e., 19,995 data points),
which may overwhelm memory limitations set in some computer modeling programs.
70
-------
o>
Data = 19995
Perc = 100.0 %
50 100
Time from injection (d)
150
Figure 41. Decimal time-concentration data file for the Fluor_time.prn data file developed from a
5 km tracer test (100% of the data shown).
Data = 140
Perc = 0.70
50 100
Time from injection (d)
150
Figure 42. Decimal time-concentration data file for the Fluor Jime.prn data file developed from a
5 km tracer test where data are processed for each day (0.71% of the data shown).
71
-------
Data = 71
Perc = 0.36 %
Od
50 100
Time from injection (d)
150
Figure 43. Decimal time-concentration data file for the Fluor_time.prn data file developed from a
5 km tracer test where data are processed for every other day (0.36% of the data shown).
o
o
20
'-• 15
o>
10
Od
Data = 29
Perc = 0.15
50 100
Time from injection (d)
150
Figure 44. Decimal time-concentration data file for the Fluor_time.prn data file developed from a
5 km tracer test where data are processed for every fifth day (0.15% of the data shown).
72
-------
Data = 15
Perc = 0.075 %
50 100
Time from injection (d)
Figure 45. Decimal time-concentration data file for the Fluor Jime.prn data file developed from a
5 km tracer test where data are processed for every tenth day (0.075% of the data shown).
Data = 400
Perc = 2.0 %
50 100
Time from injection (d)
Figure 46. Decimal time-concentration data file for the Fluor Jime.prn data file developed from a
5 km tracer test and set for 2% of the data as shown.
73
-------
5.3.2. Turner Designs Data File Break.prn
The data file Break.prn was developed from a field study in which the Model 10-AU-005 was
set to log data every 10 min with concentrations reported in |ig L"1. This data file was modified
to simulate the effect of an extremely large time break in the logged data file. Data downloading
occurred several times after tracer injection. The first 10 instances of data downloading resulted in
relatively insignificant logger interruptions during which logged data were downloaded. However,
~33 days after tracer injection, a power outage disrupted data logging. Due to the length of time
of this experiment, an equipment check was not performed until the next planned date for data
downloading (~58 days after tracer injection). The effect of logger shutdown was the loss of
~24.7 days of tracer data at a critical time during the tracer experiment.
Unlike the minor time break in the Creek.prn data set, the major time break in the Break.prn
file is very significant because so much data were lost and because it occurred just after(?) peak
tracer arrival (realistically, it is impossible to know the full effect of the missing data; it is entirely
possible that peak recovery was actually missed). However, the data that were acquired still need
to be properly assessed, which requires that the lost time be properly taken into account and the
subsequent data properly adjusted for the lost time.
Figure 47 is a plot of the time-concentration data file in decimal days using all the data. Arrows
depicting the location of the lost time (~25.5 days) are shown in Figure 47. Because there are so
much lost data in the Break.prn file, a huge break in time is shown in Figure 47.
Figure 48 is a plot of the time-concentration data file in decimal days for 1.5% of the
total logged time-concentration data (FLOWTHRU-determined percentage). The arrows shown
on Figure 48 more clearly indicates the location of missing time relative to the recorded data.
From Figure 48 (and equally so from Figure 47) it may be discerned that the break in time
recording occurred slightly after 30 days since the time of injection. (Note: The arrows depicted
in Figures 47 and 48 are shown only for illustration purposes and are not part of the FLOWTHRU
package.)
5.3.3. Turner Designs Data File 03gl58.prn
The data file 03gl58.prn was developed from a field study in which the Model 10-AU-005
developed by Michael Verrault (Les Laboratoires SL, Chicoutimi, Quebec) was set to log data
every 30 min with concentrations listed in (ig L"1. The test consisted of an injection of 0.457 kg
of dye over one hour on November 11, 2003, beginning at 1320 hours.
Unfortunately, almost immediately after the test was initiated, a power failure and subsequent
74
-------
Data = 9555
Perc = 100.0
Location of Missing
Time (~24.7 days)
40 60
Time from injection (d)
Figure 47. Decimal time-concentration data file for the Break.prn data file developed from a 5 km
tracer test (100% of the data shown). Note arrows indicating the location of missing time.
Data = 143
Perc = 1.5 %
Location of Missing
Time (~24.7 days)
40 60
Time from injection (d)
Figure 48. Decimal time-concentration data file for the Break.prn data file developed from a 5 km
tracer test (1.5% of the total data shown). Note arrows indicating the location of missing time.
75
-------
power surge severely damaged an internal NVRAM memory chip in the Model 10-AU-005,
necessitating replacement. The time delay caused by determining the source of the instrument's
internal problem, ordering a new NAVRAM chip, and installing the chip was significant. The
fluorometer was repaired and restarted on November 24, 2003, at 1158 hours resulting in a total
down time that exceeded 21 days. However, tracer migration was sufficiently slow such that a
large portion of the tracer was still detected and logged by the fluorometer.
Figure 49 is a plot of all the logged data, Figure 50 is a plot of ~50% of the data (50% chosen
by "user"), and Figure 51 is a plot of ~11% of the data (percentage determined by computer).
Arrows depicting the location of the lost time (13 days) are shown in Figures 49-51. Because
there are so much lost data in 03gl58.prn, a huge break in time is shown immediately after the
start of the tracer test in Figures 49-51.
Figure 51 illustrates a potentially serious problem that occurs when the percentage of data used
in processing FLOWTHRU is too small. Comparing Figure 51 with Figures 49 and 50 shows that a
significant amount of important data around the "peak concentration" where concentration values
were fluctuating were not processed by FLOWTHRU when only 11% of the data was processed. In
this particular instance the low percentage of data display is not a serious factor in terms of ETC
analysis (ignoring the data missing as a result of instrument failure). Comparing Figures 49, 50,
and 51 shows that even though a selection of 11% of data was used in processing Figure 51, the
unprocessed data lie almost directly below the peak concentration value.
It will also be noted that the last data point in the data file 03gl58.prn was also not processed by
FLOWTHRU when only 10% of the data were processed. As a result, the dropoff in concentration
for the final logged data value is not apparent from Figure 49. Again, this unprocessed data point
lies almost directly below the immediately prior data, so the effect of not processing this last data
value is not serious in terms of time-of-travel analyses. However, the user will need to check
FLOWTHRU output files carefully to ensure that all data that should have been processed were
actually processed.
5.3.3.1. Effect of Data Point Deletion on File 03gl58.prn. The failure of the NAVRAM chip
in the fluorometer and the eventual restarting of the fluorometer are suspected of also having
resulted in some anomalous data points. Specifically, the second data point (12.9434,0.086) is
suspect and the very last data point (21.8634,0.123) is questionable (see Figures 49-51).
Using the interactive feature of FLOWTHRU to select these two data points resulted in the
deletion of these data points. Figure 52 is a plot of the "full" data set minus the two deleted
data points. Comparing Figure 52 with Figure 49 shows a reduction in the number of data points
76
-------
'-• 4
J
|
—1—1—1—1—'—
Data = 426
Perc = 100.0%
Location of Missing
Time (~13.0 days)
10 15
Time from injection (d)
20
Figure 49. Decimal time-concentration data file for the 03gl58.prn data file developed from a
tracer test (100% of the data shown). Note arrows indicating the location of missing time.
s
g
Data = 214
Perc = 50.2 %
Location of Missing
Time (~13.0 days)
5 10 15 20
Time from injection (d)
Figure 50. Decimal time-concentration data file for the 03gl58.prn data file developed from a
tracer test (~50% of the data shown). Note arrows indicating the location of missing time.
77
-------
Data = 49
Perc = 11.5 %
2
8
Location of Missing
Time (~13.0 days)
0*-
10 15
Time from injection (d)
20
Figure 51. Decimal time-concentration data file for the 03gl58.prn data file developed from a
tracer test (~11% of the data shown). Note arrows indicating the location of missing time.
6 -
I
§
10 15
Time from injection (d)
20
Figure 52. Decimal time-concentration data file for the 03gl58.prn data file in which the second
(12.9434,0.086) and last (21.8634,0.123) data points have been deleted using FLOWTHRU.
78
-------
processed but those that were processed were counted as 100% complete.
79
-------
6. SUMMARY
Continuous recording of dye fluorescence using field fluorometers at selected sampling sites
is a commonly applied measurement technique. Unfortunately, the general form of data logged by
a Turner Designs Model 10-AU-005 and downloaded using the Turner Designs IDL_1B 1 program
does not readily lend itself to rapid and easy analysis. A new computer program, FLOWTHRU,
efficiently reads the data logged by the Model 10-AU-005, converts the data to decimal time,
and plots the data directly to a computer monitor, which allows trends in the data to be quickly
discerned. Use of FLOWTHRU on Model 10-AU-005 logged data is a useful complement to
the !DL_lBl program and facilitates data analysis. Useful features of the program FLOWTHRU
include:
1. the ability to read-in a simple process control data file or interactive responses from the user;
2. extensive error-checking and error-correction routine to facilitate data processing;
3. conversion of logged date and time to decimal days, hours, minutes, seconds;
4. processing of measured data, a percentage of the data set by the user or by FLOWTHRU,
daily or hourly data, block averaged data, or smoothed data using a moving average;
5. listing and file creation of the converted time-concentration data file for use in other
modeling packages;
6. listing and file creation of background time-concentration data and water temperature data;
7. date and time adjustments to account for lost time due to instrument malfunction or other
circumstance; and
8. graphical and publication-quality PostScript plot generation of FLOWTHRU, results allowing
for real-time evaluation of the tracer data.
This program greatly enhances the capabilities of the Turner Designs Model 10-AU-005 field
filter fluorometer when used in the continuous flow-through mode. It should lead to greater
reliance on continuous monitoring during dye-tracing experiments in the future because of the
vastly superior amount of data collected, quality (accuracy and precision) of the data collected,
and ease with which the data may be evaluated.
80
-------
APPENDIX I
TABLE OF ERRORS RESULTING FROM A COMPARISON OF MEASURED
CONCENTRATIONS VERSUS SMOOTHED CONCENTRATIONS
Sample
Number
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
Time Measured Cone.
(min) (ppm)
0.0000
0.0167
0.0334
0.0500
0.0667
0.0834
0.1000
0.1167
0.1334
0.1500
0.1667
0.1834
0.2000
0.2167
0.2334
0.2500
0.2667
0.2834
0.3000
0.3167
0.3334
0.3500
0.3667
0.3834
0.4000
0.4167
0.4334
0.4500
0.4667
0.4834
0.5000
0.5167
0.5334
0.160
0.160
0.160
0.160
0.160
0.160
0.164
0.189
0.238
0.250
0.228
0.208
0.199
0.189
0.172
0.163
0.179
0.203
0.204
0.190
0.174
0.164
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.159
0.160
0.162
0.162
Smoothed Cone.
(ppm)
0.094
0.103
0.112
0.121
0.130
0.139
0.148
0.157
0.166
0.175
0.184
0.189
0.192
0.195
0.196
0.196
0.198
0.202
0.203
0.199
0.192
0.186
0.182
0.178
0.175
0.174
0.174
0.172
0.168
0.164
0.162
0.160
0.160
Absolute Error
(dimen.)
0.066
0.057
0.048
0.039
0.030
0.021
0.016
0.032
0.072
0.075
0.044
0.019
0.007
-0.006
-0.024
-0.033
-0.019
0.001
0.001
-0.009
-0.018
-0.022
-0.022
-0.018
-0.015
-0.014
-0.014
-0.012
-0.008
-0.005
^-0.002
0.002
0.002
Relative Error
(dimen.)
(X4l3
0.356
0.300
0.243
0.187
0.130
0.096
0.168
0.301
0.298
0.191
0.092
0.033
-0.032
-0.140
-0.205
-0.107
0.007
0.005
-0.045
-0.102
-0.134
-0.135
-0.113
-0.096
-0.089
-0.087
-0.077
-0.052
-0.033
-0.010
0.010
0.011
continued on next page
81
-------
Sample
Numbei
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
Time !
(min)
0.5500
0.5667
0.5834
0.6000
0.6167
0.6334
0.6500
0.6667
0.6834
0.7000
0.7167
0.7334
0.7500
0.7667
0.7834
0.8000
0.8167
0.8334
0.8500
0.8667
0.8834
0.9000
0.9167
0.9334
0.9500
0.9667
0.9834
1.0000
.0167
.0334
.0500
.0667
.0834
.1000
.1167
.1334
.1500
.1667
1.1834
1.2000
Measured Cone.
(ppm)
0.162
0.160
0.159
0.159
0.160
0.160
0.162
0.160
0.160
0.160
0.160
0.160
0.159
0.160
0.159
0.158
0.158
0.159
0.160
0.162
0.162
0.160
0.160
0.160
0.162
0.160
0.160
0.160
0.160
0.160
0.158
0.158
0.158
0.160
0.160
0.160
0.159
0.159
0.159
0.159
Smoothed Cone
(ppm)
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.161
0.161
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.161
0.161
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.159
0.159
Absolute Error
(dimen.)
0.002
0.000
-0.001
-0.001
0.000
0.000
0.002
-0.001
-0.001
0.000
0.000
0.000
-0.001
0.000
-0.001
-0.002
-0.002
0.000
0.001
0.002
0.002
0.000
0.000
0.000
0.002
0.000
0.000
0.000
-0.001
-0.001
-0.002
-0.002
-0.002
0.000
0.000
0.001
0.000
0.000
0.000
0.000
Relative Error
(dimen.)
0.010
-0.003
-0.009
-0.008
-0.002
-0.002
0.010
-0.003
-0.003
-0.002
-0.001
0.000
-0.006
0.000
-0.006
-0.012
-0.010
-0.002
0.004
0.015
0.014
0.002
0.002
0.001
0.012
-0.001
-0.002
-0.003
-0.003
-0.003
-0.014
-0.012
-0.010
0.002
0.002
0.003
-0.002
-0.002
-0.001
-0.001
continued on next page
82
-------
Sample
Number
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
Time Measured Cone.
(min) (ppm)
1.2167
1.2334
1.2500
1.2667
1.2834
1.3000
1.3167
1.3334
1.3500
1.3667
1.3834
1.4000
1.4167
1.4334
1.4500
1.4667
1.4834
1.5000
1.5167
1.5334
1.5500
1.5667
1.5834
1.6000
1.6167
1.6334
1.6500
1.6667
1.6834
1.7000
1.7167
1.7334
1.7500
1.7667
1.7834
1.8000
1.8167
1.8334
1.8500
1.8667
0.159
0.159
0.159
0.159
0.159
0.159
0.160
0.159
0.159
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.162
0.163
0.163
0.162
0.160
0.160
0.160
0.159
0.160
0.160
0.160
0.159
0.159
0.162
0.162
0.160
0.160
0.160
0.162
0.160
0.159
0.160
0.160
Smoothed Cone.
(ppm)
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.160
0.160
0.160
0.160
0.160
0.160
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
Absolute Error
(dimen.)
0.000
0.000
0.000
0.000
0.000
0.000
0.001
0.000
0.000
0.001
0.001
0.001
0.001
0.000
0.000
0.000
0.000
0.002
0.003
0.002
0.001
-0.001
-0.001
-0.001
-0.002
-0.001
-0.001
-0.001
-0.002
-0.001
0.002
0.002
0.000
0.000
0.000
0.002
0.000
-0.001
0.000
0.000
Relative Error
(dimen.)
0.000
-0.001
-0.001
-0.002
-0.001
-0.001
0.006
-0.001
-0.001
0.005
0.005
0.004
0.003
0.003
0.002
0.002
0.001
0.012
0.016
0.014
0.007
-0.006
-0.006
-0.006
-0.011
-0.005
-0.005
-0.005
-0.010
-0.007
0.012
0.012
-0.001
-0.001
-0.001
0.010
-0.002
-0.008
-0.002
-0.002
continued on next page
83
-------
Sample Time Measured Cone.
Number (min) (ppm)
Fl4
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
142
143
144
145
146
147
148
149
150
151
152
153
1.8834
1.9000
1.9167
1.9334
1.9500
1.9667
1.9834
2.0000
2.0167
2.0334
2.0500
2.0667
2.0834
2.1000
2.1167
2.1334
2.1500
2.1667
2.1834
2.2000
2.2167
2.2334
2.2500
2.2667
2.2834
2.3000
2.3167
2.3334
2.3500
2.3667
2.3834
2.4000
2.4167
2.4334
2.4500
2.4667
2.4834
2.5000
2.5167
2.5334
0.159
0.159
0.160
0.162
0.162
0.162
0.162
0.160
0.162
0.162
0.163
0.165
0.165
0.165
0.166
0.168
0.170
0.170
0.172
0.173
0.175
0.178
0.179
0.179
0.181
0.182
0.184
0.187
0.187
0.187
0.188
0.190
0.193
0.194
0.195
0.195
0.197
0.200
0.203
0.206
Smoothed Cone.
(ppm)
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.161
0.161
0.161
0.162
0.162
0.163
0.163
0.164
0.164
0.165
0.166
0.167
0.168
0.170
0.171
0.172
0.174
0.175
0.177
0.178
0.180
0.181
0.182
0.184
0.185
0.187
0.188
0.189
0.191
0.192
0.194
0.195
Absolute Error
(dimen.)
-0.001
-0.001
0.000
0.002
0.002
0.002
0.002
0.000
0.001
0.001
0.002
0.003
0.003
0.002
0.003
0.004
0.006
0.005
0.006
0.006
0.007
0.008
0.008
0.007
0.007
0.007
0.007
0.009
0.007
0.006
0.006
0.006
0.008
0.007
0.007
0.006
0.006
0.008
0.009
0.011
Relative Error
(dimen.)
-0.009
-0.007
0.001
0.012
0.011
0.010
0.010
-0.003
0.008
0.007
0.011
0.020
0.017
0.014
0.018
0.026
0.033
0.029
0.034
0.034
0.038
0.046
0.045
0.038
0.040
0.037
0.040
0.047
0.039
0.032
0.029
0.033
0.041
0.038
0.036
0.029
0.032
0.040
0.047
0.052
continued on next page
84
-------
Sample
Number
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
Time Measured Cone.
(min) (ppm)
2.5500
2.5667
2.5834
2.6000
2.6167
2.6334
2.6500
2.6667
2.6834
2.7000
2.7167
2.7334
2.7500
2.7667
2.7834
2.8000
2.8167
2.8334
2.8500
2.8667
2.8834
2.9000
2.9167
2.9334
2.9500
2.9667
2.9834
3.0000
3.0167
3.0334
3.0500
3.0667
3.0834
3.1000
3.1167
3.1334
3.1500
3.1667
3.1834
3.2000
0.206
0.208
0.209
0.209
0.209
0.209
0.210
0.209
0.209
0.211
0.212
0.214
0.212
0.212
0.214
0.211
0.210
0.210
0.211
0.211
0.209
0.209
0.209
0.209
0.209
0.207
0.204
0.204
0.203
0.203
0.202
0.200
0.199
0.196
0.195
0.196
0.195
0.194
0.193
0.189
Smoothed Cone.
(ppm)
0.197
0.199
0.201
0.202
0.203
0.205
0.206
0.207
0.208
0.209
0.209
0.210
0.210
0.211
0.211
0.211
0.211
0.211
0.211
0.212
0.211
0.211
0.211
0.210
0.210
0.210
0.209
0.208
0.208
0.207
0.206
0.205
0.204
0.203
0.202
0.201
0.200
0.199
0.198
0.197
Absolute Error
(dimen.)
0.009
0.009
0.008
0.007
0.006
0.004
0.004
0.002
0.001
0.002
0.003
0.004
0.002
0.001
0.003
0.000
-0.001
-0.001
0.000
-0.001
-0.002
-0.002
-0.002
-0.001
-0.001
-0.003
-0.005
-0.004
-0.005
-0.004
-0.004
-0.005
-0.005
-0.007
-0.007
-0.005
-0.005
-0.005
-0.005
-0.008
Relative Error
(dimen.)
0.044
0.044
0.040
0.033
0.027
0.021
0.019
0.009
0.005
0.011
0.013
0.019
0.008
0.007
0.014
-0.001
-0.006
-0.006
-0.002
-0.003
-0.012
-0.010
-0.008
-0.007
-0.006
-0.012
-0.024
-0.021
-0.023
-0.020
-0.021
-0.027
-0.027
-0.037
-0.036
-0.025
-0.024
-0.025
-0.025
-0.040
continued on next page
85
-------
Sample Time Measured Cone.
Number (min) (ppm)
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
3.2167
3.2334
3.2500
3.2667
3.2834
3.3000
3.3167
3.3334
3.3500
3.3667
3.3834
3.4000
3.4167
3.4334
3.4500
3.4667
3.4834
3.5000
3.5167
3.5334
3.5500
3.5667
3.5834
3.6000
3.6167
3.6334
3.6500
3.6667
3.6834
3.7000
3.7167
3.7334
3.7500
3.7667
3.7834
3.8000
3.8167
3.8334
3.8500
3.8667
0.189
0.189
0.187
0.187
0.187
0.186
0.186
0.186
0.185
0.181
0.180
0.179
0.179
0.177
0.175
0.175
0.175
0.173
0.173
0.172
0.170
0.170
0.170
0.170
0.170
0.170
0.170
0.170
0.167
0.166
0.167
0.166
0.165
0.165
0.165
0.165
0.165
0.165
0.165
0.165
Smoothed Cone
(ppm)
0.195
0.194
0.193
0.192
0.191
0.190
0.189
0.188
0.188
0.187
0.186
0.185
0.184
0.183
0.182
0.181
0.180
0.179
0.177
0.176
0.175
0.174
0.174
0.173
0.172
0.172
0.171
0.171
0.170
0.170
0.169
0.169
0.168
0.168
0.167
0.167
0.166
0.166
0.166
0.165
. Absolute Error
(dimen.)
-0.006
-0.005
-0.006
-0.005
-0.004
-0.004
-0.003
-0.002
-0.003
-0.006
-0.006
-0.006
-0.005
-0.006
-0.007
-0.006
-0.005
-0.006
-0.004
-0.004
-0.005
-0.004
-0.004
-0.003
-0.002
-0.002
-0.001
-0.001
-0.003
-0.004
-0.002
-0.003
-0.003
-0.003
-0.002
-0.002
-0.001
-0.001
-0.001
0.000
Relative Error
(dimen.)
-0.033
-0.027
-0.032
-0.026
-0.021
-0.022
-0.018
-0.013
-0.014
-0.031
-0.032
-0.033
-0.027
-0.034
-0.039
-0.033
-0.028
-0.033
-0.026
-0.025
-0.031
-0.026
-0.021
-0.016
-0.012
-0.010
-0.007
-0.004
-0.019
-0.021
-0.013
-0.016
-0.020
-0.017
-0.014
-0.012
-0.009
-0.006
-0.003
-0.002
continued on next page
86
-------
Sample
Number
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
Time Measured Cone.
(min) (ppm)
3.8834
3.9000
3.9167
3.9334
3.9500
3.9667
3.9834
4.0000
4.0167
4.0334
4.0500
4.0667
4.0834
4.1000
4.1167
4.1334
4.1500
4.1667
4.1834
4.2000
4.2167
4.2334
4.2500
4.2667
4.2834
4.3000
4.3167
4.3334
4.3500
4.3667
4.3834
4.4000
4.4167
4.4334
4.4500
4.4667
4.4834
4.5000
4.5167
4.5334
0.165
0.165
0.165
0.165
0.165
0.165
0.165
0.164
0.163
0.164
0.163
0.160
0.162
0.162
0.162
0.160
0.160
0.160
0.162
0.164
0.164
0.164
0.163
0.163
0.163
0.164
0.162
0.159
0.158
0.160
0.160
0.160
0.160
0.162
0.162
0.160
0.160
0.160
0.159
0.158
Smoothed Cone.
(ppm)
0.165
0.165
0.165
0.165
0.165
0.165
0.165
0.165
0.165
0.165
0.164
0.164
0.164
0.163
0.163
0.163
0.162
0.162
0.162
0.162
0.162
0.162
0.162
0.162
0.162
0.162
0.163
0.163
0.162
0.162
0.162
0.161
0.161
0.161
0.161
0.161
0.160
0.160
0.160
0.160
Absolute Error
(dimen.)
0.000
0.000
0.000
0.000
0.000
0.000
0.000
-0.001
-0.002
-0.001
-0.001
-0.004
-0.002
-0.001
-0.001
-0.003
-0.002
-0.002
0.000
0.002
0.002
0.002
0.001
0.001
0.001
0.002
-0.001
-0.004
-0.004
-0.002
-0.002
-0.001
-0.001
0.001
0.001
-0.001
0.000
0.000
-0.001
-0.002
Relative Error
(dimen.)
-0.002
-0.001
0.000
0.000
0.000
0.000
0.000
-0.006
-0.011
-0.004
-0.009
-0.025
-0.011
-0.009
-0.007
-0.017
-0.014
-0.011
0.002
0.014
0.014
0.013
0.006
0.005
0.004
0.009
-0.004
-0.022
-0.028
-0.014
-0.011
-0.009
-0.007
0.006
0.007
-0.004
-0.002
-0.001
-0.007
-0.013
continued on next page
87
-------
Sample Time Measured Cone.
Number (min) (ppm)
274 r~~~
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
4.5500
4.5667
4.5834
4.6000
4.6167
4.6334
4.6500
4.6667
4.6834
4.7000
4.7167
4.7334
4.7500
4.7667
4.7834
4.8000
4.8167
4.8334
4.8500
4.8667
4.8834
4.9000
4.9167
4.9334
4.9500
4.9667
4.9834
5.0000
5.0167
5.0334
5.0500
5.0667
5.0834
5.1000
5.1167
5.1334
5.1500
5.1667
5.1834
5.2000
0.159
0.160
0.160
0.160
0.160
0.159
0.159
0.158
0.158
0.159
0.159
0.158
0.158
0.159
0.159
0.160
0.160
0.160
0.162
0.160
0.159
0.160
0.160
0.159
0.158
0.158
0.159
0.159
0.160
0.160
0.160
0.159
0.159
0.159
0.160
0.162
0.162
0.160
0.160
0.160
Smoothed Cone.
(ppm)
0.160
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.160
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.160
0.160
0.160
0.160
0.160
Absolute Error
(dimen.)
-0.001
0.000
0.000
0.000
0.000
-0.001
0.000
-0.001
-0.001
0.000
0.000
-0.001
-0.001
0.000
0.000
0.001
0.001
0.001
0.003
0.001
0.000
0.000
0.000
-0.001
-0.002
-0.002
-0.001
0.000
0.001
0.001
0.001
0.000
0.000
0.000
0.001
0.002
0.002
0.000
0.000
0.000
Relative Error
(dimen.)
-0.006
0.000
0.000
0.000
0.001
-0.003
-0.003
-0.008
-0.007
-0.001
-0.001
-0.007
-0.006
0.001
0.002
0.008
0.007
0.007
0.017
0.003
-0.003
0.003
0.002
-0.005
-0.011
-0.010
-0.003
-0.003
0.003
0.005
0.005
-0.002
-0.001
-0.001
0.005
0.015
0.013
0.000
-0.001
-0.001
continued on next page
88
-------
Sample
Number
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
Time Measured Cone.
(min) (ppm)
5.2167
5.2334
5.2500
5.2667
5.2834
5.3000
5.3167
5.3334
5.3500
5.3667
5.3834
5.4000
5.4167
5.4334
5.4500
5.4667
5.4834
5.5000
5.5167
5.5334
5.5500
5.5667
5.5834
5.6000
5.6167
5.6334
5.6500
5.6667
5.6834
5.7000
5.7167
5.7334
5.7500
5.7667
5.7834
5.8000
5.8167
5.8334
5.8500
5.8667
0.160
0.159
0.159
0.160
0.162
0.162
0.162
0.160
0.160
0.160
0.162
0.162
0.162
0.162
0.160
0.160
0.162
0.160
0.160
0.162
0.162
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.158
0.158
0.158
0.158
0.159
0.160
0.162
0.159
0.158
0.159
0.159
0.158
Smoothed Cone.
(ppm)
0.160
0.160
0.160
0.160
0.160
0.161
0.161
0.160
0.160
0.160
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
Absolute Error
(dimen.)
0.000
-0.001
-0.001
0.000
0.002
0.001
0.001
0.000
0.000
0.000
0.001
0.001
0.001
0.001
-0.001
-0.001
0.001
-0.001
-0.001
0.001
0.001
-0.001
-0.001
-0.001
-0.001
-0.001
-0.001
-0.001
-0.002
-0.002
-0.001
-0.001
0.000
0.001
0.003
0.000
-0.001
0.000
0.000
-0.001
Relative Error
(dimen.)
-0.001
-0.006
-0.006
-0.001
0.010
0.009
0.009
-0.002
-0.002
-0.002
0.009
0.008
0.006
0.004
-0.008
-0.007
0.006
-0.006
-0.006
0.006
0.004
-0.007
-0.006
-0.005
-0.003
-0.003
-0.009
-0.007
-0.013
-0.012
-0.009
-0.007
0.000
0.006
0.017
-0.001
-0.006
0.001
0.001
-0.006
continued on next page
89
-------
Sample Time Measured Cone.
Number (min) (ppm)
354 ~
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
5.8834
5.9000
5.9167
5.9334
5.9500
5.9667
5.9834
6.0000
6.0167
6.0334
6.0500
6.0667
6.0834
6.1000
6.1167
6.1334
6.1500
6.1667
6.1834
6.2000
6.2167
6.2334
6.2500
6.2667
6.2834
6.3000
6.3167
6.3334
6.3500
6.3667
6.3834
6.4000
6.4167
6.4334
6.4500
6.4667
6.4834
6.5000
6.5167
6.5334
0.159
0.160
0.160
0.160
0.159
0.160
0.162
0.162
0.160
0.159
0.159
0.159
0.159
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.160
0.159
0.159
0.160
0.160
0.160
0.160
0.160
0.159
0.158
0.158
0.159
0.159
0.159
0.160
0.160
0.160
0.160
0.159
0.159
Smoothed Cone.
(ppm)
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.159
0.159
0.160
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
Absolute Error
(dimen.)
0.000
0.001
0.001
0.001
0.000
0.001
0.003
0.002
0.000
-0.001
-0.001
-0.001
-0.001
0.000
0.000
0.000
0.000
-0.001
0.000
0.000
0.001
0.000
0.000
0.000
0.000
0.000
0.000
0.000
-0.001
-0.001
-0.001
0.000
0.000
0.000
0.001
0.001
0.001
0.001
0.000
0.000
Relative Error
(dimen.)
0.000
0.005
0.004
0.003
-0.002
0.005
0.016
0.013
0.001
-0.006
-0.006
-0.006
-0.006
0.001
0.001
0.000
0.000
-0.005
-0.003
-0.002
0.003
-0.003
-0.003
0.003
0.003
0.003
0.003
0.003
-0.003
-0.009
-0.009
-0.002
-0.002
-0.002
0.005
0.005
0.005
0.005
-0.001
-0.001
continued on next page
90
-------
Sample
Number
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
Time Measured Cone.
(min) (ppm)
6.5500
6.5667
6.5834
6.6000
6.6167
6.6334
6.6500
6.6667
6.6834
6.7000
6.7167
6.7334
6.7500
6.7667
6.7834
6.8000
6.8167
6.8334
6.8500
6.8667
6.8834
6.9000
6.9167
6.9334
6.9500
6.9667
6.9834
7.0000
7.0167
7.0334
7.0500
7.0667
7.0834
7.1000
7.1167
7.1334
7.1500
7.1667
7.1834
7.2000
0.160
0.160
0.158
0.158
0.159
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.159
0.159
0.160
0.159
0.158
0.159
0.158
0.158
0.158
0.158
0.159
0.159
0.160
0.160
0.160
0.160
0.162
0.160
0.162
0.163
0.162
0.159
0.159
0.160
0.160
0.160
0.160
0.160
Smoothed Cone.
(ppm)
0.159
0.160
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.160
0.160
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.160
Absolute Error
(dimen.)
0.001
0.000
-0.001
-0.001
0.000
0.001
0.001
0.001
0.001
0.000
0.000
0.000
0.000
0.000
0.001
0.000
-0.001
0.000
-0.001
-0.001
-0.001
-0.001
0.000
0.000
0.001
0.001
0.001
0.001
0.003
0.001
0.002
0.003
0.001
-0.002
-0.002
-0.001
-0.001
-0.001
-0.001
0.000
Relative Error
(dimen.)
0.004
0.003
-0.009
-0.009
-0.002
0.004
0.004
0.004
0.004
-0.002
-0.002
-0.002
-0.001
-0.002
0.003
-0.003
-0.008
-0.001
-0.006
-0.005
-0.005
-0.004
0.002
0.002
0.008
0.008
0.007
0.006
0.017
0.003
0.013
0.017
0.008
-0.010
-0.010
-0.004
-0.004
-0.004
-0.004
-0.003
continued on next page
91
-------
Sample Time Measured Cone.
Number (min) (ppm)
434 =
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
7.2167
7.2334
7.2500
7.2667
7.2834
7.3000
7.3167
7.3334
7.3500
7.3667
7.3834
7.4000
7.4167
7.4334
7.4500
7.4667
7.4834
7.5000
7.5167
7.5334
7.5500
7.5667
7.5834
7.6000
7.6167
7.6334
7.6500
7.6667
7.6834
7.7000
7.7167
7.7334
7.7500
7.7667
7.7834
7.8000
7.8167
7.8334
7.8500
7.8667
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.159
0.160
0.160
0.159
0.160
0.160
0.160
0.160
0.162
0.162
0.160
0.160
0.159
0.159
0.158
0.158
0.158
0.159
Smoothed Cone
(ppm)
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
. Absolute Error
(dimen.)
0.000
0.000
0.000
0.000
0.000
0.000
0.000
-0.001
-0.001
-0.001
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
-0.001
-0.001
-0.001
-0.001
0.000
0.000
-0.001
0.000
0.000
0.000
0.000
0.002
0.002
0.000
0.000
-0.001
-0.001
-0.002
-0.002
-0.002
-0.001
Relative Error
(dimen.)
-0.003
-0.002
0.000
0.001
0.001
0.000
0.000
-0.006
-0.005
-0.005
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.001
-0.005
-0.005
-0.005
-0.004
0.002
0.002
-0.003
0.003
0.003
0.003
0.003
0.013
0.012
-0.001
-0.002
-0.007
-0.007
-0.013
-0.012
-0.010
-0.003
continued on next page
92
-------
Sample
Number
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
Time Measured Cone.
(min) (ppm)
7.8834
7.9000
7.9167
7.9334
7.9500
7.9667
7.9834
8.0000
8.0167
8.0334
8.0500
8.0667
8.0834
8.1000
8.1167
8.1334
8.1500
8.1667
8.1834
8.2000
8.2167
8.2334
8.2500
8.2667
8.2834
8.3000
8.3167
8.3334
8.3500
8.3667
8.3834
8.4000
8.4167
8.4334
8.4500
8.4667
8.4834
8.5000
8.5167
8.5334
0.159
0.159
0.158
0.158
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.160
0.160
0.160
0.159
0.158
0.158
0.159
0.160
0.160
0.159
0.158
0.159
0.160
0.159
0.160
0.162
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
Smoothed Cone.
(ppm)
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.159
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
Absolute Error
(dimen.)
0.000
0.000
-0.001
-0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.000
0.000
0.000
-0.001
-0.001
0.000
0.000
0.000
-0.001
-0.002
-0.001
0.000
0.001
0.001
0.000
-0.001
0.000
0.001
0.000
0.001
0.003
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
Relative Error
(dimen.)
-0.003
-0.001
-0.005
-0.004
0.009
0.008
0.007
0.006
0.005
0.004
0.003
0.003
0.002
0.001
-0.006
-0.005
0.001
0.001
0.001
-0.005
-0.010
-0.009
-0.002
0.005
0.005
-0.002
-0.007
-0.001
0.006
0.000
0.006
0.016
0.002
0.002
0.002
0.002
0.001
0.000
-0.001
-0.001
continued on next page
93
-------
Sample Time Measured Cone.
Number (min) (ppm)
514 :
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
8.5500
8.5667
8.5834
8.6000
8.6167
8.6334
8.6500
8.6667
8.6834
8.7000
8.7167
8.7334
8.7500
8.7667
8.7834
8.8000
8.8167
0.160
0.160
0.160
0.160
0.160
0.160
0.162
0.162
0.162
0.162
0.162
0.162
0.160
0.160
0.160
0.162
0.163
Smoothed Cone.
(ppm)
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.160
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.161
0.162
Absolute Error
(dimen.)
0.000
0.000
0.000
0.000
0.000
0.000
0.002
0.002
0.001
0.001
0.001
0.001
-0.001
-0.001
-0.001
0.001
0.001
Relative Error
(dimen.)
-0.001
-0.001
0.000
0.000
0.000
0.000
0.011
0.010
0.009
0.008
0.007
0.006
-0.007
-0.007
-0.007
0.004
0.009
94
-------
References
ACCC, 1998. ISO 8601 dates: What they are and how to use them; ISO 8601 dates: What
they are and why they're good; Appendix 2: All about leap years. Tech. Rep. Univ. of
Illinois at Chicago, Academic Computing and Communications Center, http://www.
uic.edu/depts/accc/sof tware/isodates/leapyear.htmlttrules, Chicago,
111., [accessed 14 August 2003].
Brandt, S., 1999. Data Analysis: Statistical and Computational Methods for Scientists and
Engineers. Springer-Verlag, New York.
Field, M. S., 2002. The QTRACER2 program for tracer-breakthrough curve analysis for hydro-
logical tracer tests. Tech. Rep. EPA/600/R-02/001, U.S. Envir. Prot. Agency, Washington, D.C.
Lin, A. Y.-C., Debroux, J.-F., Cunningham, J. A., Reinhard, M., 2003. Comparison of rhodamine
WT and bromide in the determination of hydraulic characteristics of constructed wetlands.
Ecological Engineering 20, 75-88.
Mull, D. S., Liebermann, T. D., Smoot, J. L., Woosley, Jr., L. H., 1988. Application of Dye-
Tracing Techniques for Determining Solute-Transport Characteristics of Ground Water in Karst
Terranes. Tech. Rep. EPA/904/9-88-001, 103 p., U.S. Envir. Prot. Agency, Region IV, Atlanta,
Ga.
Pearson, T. J., 1997. Pgplot graphics subroutine library. Tech. rep., California Institute of Tech-
nology, Pasadena.
Smart, C. C, Zabo, L., Alexander, Jr., E. C., Worthington, S. R. H., 1998. Some advances in
fluorometric techniques for water tracing. Environmental Monitoring and Assessment 53, 305-
320.
Toride, N., Leij, F. J., van Genuchten, M. T., 1995. The CXTFIT code for estimating transport
parameters from the laboratory or field tracer experiments; version 2.0. Tech. Rep. 137, U.S.
Salinity Lab., Riverside, Calif.
Turner, D., 1993. Model 10-AU-005 field fluorometer user's manual. Tech. Rep. 10-AU-075,
Turner Designs, Sunnyvale, Calif.
95
-------
Wilson, Jr., J. F., Cobb, E. D., Kilpatrick, F. A., 1986. Fluorometric Procedures for Dye Tracing.
Tech. Rep. 34 Techniques of Water-Resources Investigations of the U.S. Geological Survey,
Book 3, Chapter A12, 34 p., U.S. Geological Survey.
96
------- |