A National Database of Travel Time, Dispersion and
Contents
1. V oo oe oe oe 0 8 0 6 Velocity m s 0 4 g 96 0 2 gt 9 4 A wo 06 6 60999 A 20 30 40 50 70 80 Discharge m3 s Fig 6 2 Variation of Mean Velocity with Discharge o An analysis similar to that performed by Jobson 1997 has been undertaken using the data available in the EA database The resulting optimised relationship has been used to compare predicted peak velocities with measured peak velocities This is shown in Fig 6 4 and produces an R value of 0 76 indicating a slightly better fit than that derived by Jobson 1997 for the US data This clearly shows the potential benefit of the analysis however determining the working limits of such a relationship requires further investigation especially 0 006 0 008 0 012 Reach Slope Fig 6 3 Variation of Mean Velocity with Reach Slope as the relationships derived for the US and EA data are not similar R amp D TECHNICAL REPORT P346 16 N co Optimised Cmax Velocity m s P O N 0 0 2 0 4 0 6 0 8 1 1 2 Predicted Cmax Velocity after Jobson 1997 m s Fig 6 4 Comparison of Predicted Peak Velocities Using Jobson 1997 Technique 6 2 Velocity Discharge Trends In reviewing current practise within the Agency Scoping Report Project Record Appendix 4 0 it was found that predictin2. Coefficient B Coefficient B Matrix 1 Coefficient A Matrix 2 6 84 6 88 6 92 Coefficient B Matrix 3 Fig 4 1 Representation of the Matrix Optimisation Procedure For the optimisation technique to be reliable it was essential to verify that there was a unique pair of coefficients which gave the best fit of the prediction to the measured data This was determined by examining the R values over the entire matrix and ensuring that there was only one peak value Fig 4 2 displays the results of this procedure for the ADZ model The base of the chart in the figure 1s composed of the travel time and time delay values used in the matrix The three dimensional surface represents the R values for each travel time and time R amp D TECHNICAL REPORT P346 11 delay combination on the base Fig 4 2a shows a very coarse matrix and it is clear that there is a single peak Fig 4 2b shows results from a more refined matrix and indicates for this sample data that the prediction is more sensitive to time delay than travel time Coefficient Travel time steps Dispersion coefficient steps 0 001 m s Time delay steps data time step At Table 4 1 Ultimate matrix resolution values LY Hn HALL gt 0 9 bx S N
3. w 9 Rhe es ZA s we perenne o q 99 of 2184 4 SO OO Oo oO 6 oe o Fig 6 8 Variation of Dispersive Fraction with Discharge o 6 30 Discharge m3 s 80 The magnitude of the dispersive fraction appears to vary less with discharge than the longitudinal dispersion coefficient and as a result it may introduce less variations if this technique were adopted for predictions However with the quality and quantity of data R amp D TECHNICAL REPORT P346 19 available to date use of different mean velocities for first arrival 102 and end of the trace C 19 Would provide a robust method to predict the spread of soluble material Dispersive Fraction R amp D TECHNICAL REPORT P346 0 9 0 8 0 7 4 o o O IN 69 A 4 oy 0 1 oe 100 150 200 250 300 350 400 450 Dispersion Coefficient m2 s Fig 6 9 Variation of Dispersion Coefficient with Dispersive Fraction 20 7 CONCLUSIONS All available temporal concentration distributions obtained from recent river tracing studies have been collected and an initial quality assurance check undertaken The acceptable data have been stored in an agreed MS Excel format containing summary detai
4. k Slope 8 Four Traces 0000002 Four Traces 0 0004 Table 6 2 Parameters Derived from Fitting v kQ Trends 6 3 Longitudinal Dispersion Trends Quality and completeness of temporal concentration distributions have a greater influence on the prediction of dispersion parameters than on the prediction of mean velocities Optimisation of both ADE and ADZ predictions have been made for the available data sets and the results are summarised below No attempt has been made at present to filter or remove any values as this becomes subjective and requires detailed consideration of all the predictions The variation of the longitudinal coefficient ADE and the dispersive fraction ADZ with discharge have been produced in Fig 6 7 and 6 8 respectively R amp D TECHNICAL REPORT P346 18 450 gt 400 350 N o Dispersion Coefficient m2 s 100 50 200 150 o o o g x Qos Fe YA sn yt 10 20 30 Fig 6 7 Variation of Dispersion Coefficient with Discharge Discharge m3 s 80 As with the travel time results on first inspection there are no clear trends and further interpretation of the data set is required It is however encouraging to note that Fig 6 9 shows a positive correlation between the longitudinal dispersion coefficient and the dispersive fraction 0 9 0 8 0 7 o o o O P Dispersive Fraction
5. q Buy uq s xy A 9u02 120110 HEYI 014 660 2 0 08 MOY ULaw PHH 1 y ulu b uonenuauo3 bulsoq buisog 11920 22 aweu aSNoNAJeA AIAH uon lul ajqeaijdde ji 3aWwey yseay Aobaje moj UIEN pua wye aly uoibay 3WEN 9114 uoneuo ul je12u35 Ce 9724 LAOdHAA TVOINHOAL AVA s noH Ou 091 Ort 021 001 08 09 02 0 G0 3 s 2 O mope wdiys IMO M o INN weypeoN we a amp er ce Hb CE 9vtd L4OdHd TV O NH HL APA FLEEzO O 92 01 900 26 0 alin 9 2000 0 92 02 9 9 157020 10 201 IZ 89000 0 z Sr6 00 0 99 952400 5 201 8899 2 0 9 0 9100 70 5 62 929 PZZEL0 O 9201 6992 901 9 Z000 0 962 9 299910 0 92 201 996022 0 92 Oz 3 2000 0 GZ 62 Gas 952400 z l 8 20610 02 9 2000 0 62 6200 0 99 15702010 Sz LOL 2289 0 5 69 289000 0 9 82 Sr6 00 0 929 FLEEZO O 9101 6921710 5969 3 2000 0 982 601000 0 9 L619z0 0 92 LOL 2212 0 92 69 289000 0 9282 6010000 ey GLZZ0 0 LOL 967960 0 69 9 2000 0 82 601000 0 Gt GLZZ0 0 2001 FE0620 0 92 99 9 2000 0 GLZ 601000 0 92 22006010 9001 2221900 989 2000
6. 0 912 t 220060 0 92 001 9289 93 2000 0 GZ zz 601000 0 GLE E98EE0 0 001 28617010 9289 9 2000 0 22 98000 0 G E 2289 0 0 S266 Z6re0 O 89 3 2000 0 sz z 601000 0 926 2 G 66 9916Z0 0 9 129 289000 0 992 601000 0 L6L9z0 0 GZ 66 20 0 9129 289000 0 9292 601000 0 9 2 EV Z 66 22861010 92 29 289000 0 92 98000 0 92 PZZELO O 62 86 GL900 0 29 ZrSL00 0 GA 890100 0 922 290900 0 5986 22 9 99 99100 0 992 22020010 2 59 1200 0 9286 129800 10 9199 289000 0 9292 9 1 29210010 86 Sz9r00 0 9299 689000 0 92 228900 10 91 862000 0 GZ 26 e S000 0 99 289000 0 GZ 92 8990100 921 0 5 6 92 99 9 Z000 0 GYZ 9 57 100 6n s noH 1 s noH 1 s noH 1 s noH uocnunus3uo aAq uon lul asuls uonanusqum Aq uonsalui asuls uonaunus3uo Aq uon lul 15 2 uon lul asuls aul mopeaudrys uones Buuonuoy TUMEONA uonu s Buuonuoy THA uonus Buonuoy Pia UONeIS buuonuoy ways Beg mey 10 3 Access database user manual Time of Travel Database N T Tunstill 1999 R amp D TECHNICAL REPORT P346 34 Before running the database This database requires Microsoft Access 97 or later Running the database directly from the disc it is supplied on is not recommended Before running the database it is recommended that you copy the database file named TOT mde onto
7. 06 4 12 0 0004 4 4 1 1 oy 0 04 bol N 2 0 0002 5 1097 0 02 4 Pops 5 S S N 400 600 800 0 00 75 95 115 135 155 175 195 215 235 255 0 0002 J Time a Incomplete Trace recording ceased early b Noisy Ill Defined Distributions 0 06 gt 0 25 F 5 T nnm 5 021 5 0 08 o 8 nmm m n 0 15 0 02 0 14 a m m ee m E 0 01 0 05 4 es ux Sf 0 r 1 100000 15000 2000 25000 30000 3500 4000 45000 50000 550000 Time 1 1 Se 08 35 09 35 10 35 11 35 12 35 13 35 14 35 15 35 16 35 17 35 18 35 19 35 20 35 21 35 22 35 23 35 im c Variable Background Values d Inadequate Concentration Resolution 7 09 00 10 00 11 00 12 00 13 00 14 00 15 00 16 00 17 00 18 00 19 00 20 00 21 00 22 00 Ti e Variable Sampling Rate d Attenuated Distributions Time Fig 3 1 Examples of Poor Quality Data After discussions at meetings of the National Project Technical sub group a format for storing all necessary raw data was agreed minutes of meeting May 98 Initially all river tracing data was converted to Excel work book format containing the following sheets Sheet 1 Summary Containing all details of the data set Sheet 2 Data Containing actual data for each measurement station In paired columns of time concentration data Sheet 3 Complot Composite time concentr
8. be achieved However basic variations of mean travel time and dispersion have been obtained 6 1 Travel Times A histogram illustrating the distribution of mean velocities is given in Fig 6 1 Frequency 0 0 125 0 25 0 375 0 5 0 625 0 75 0 875 1 1 125 Velocity m s Fig 6 1 Histogram of Reach Mean Velocities The velocities used to compile this distribution have been quantified in Table 6 1 below A total of four velocities have been obtained for both the ADE and ADZ method of analysis providing the mean travel velocity together with the velocity for the peak concentration and the start and end 10 concentrations In addition the velocities obtained from analysis of the raw data have also been presented For each method the maximum and minimum values are tabulated to illustrate the range of values obtained together with the value of the mean the variance and number of results used in the analysis R amp D TECHNICAL REPORT P346 14 Velocity m s Parameter Mean C10 55 C410 ADE 1 92 eee Variance 004 006 005 0 04 008 Raw Data Minimum 002 0 02 002 Maximum 115 105 100 Mean 031 028 025 Variance 006 0 05 0 04 Number of Values 194 194 Table 6 1 Summary of Reach Mean Velocities It is encouraging to note that there is little difference between the mean velocities obtained from both ADE and ADZ analyses and with the peak concentration velocities Cmax The
9. peak concentration It is expected that unless there are strong reasons for ending the monitoring early the tail should recede to 5 of the peak and ideally to background 4 The temporal resolution of the data should be constant at all stations and should provide a minimum of 40 points to define the distribution 5 If the EA concludes that the data is not acceptable the contractor will redo the whole trace at their expense 6 The information indicated on the EXCEL template should be provided for each survey R amp D TECHNICAL REPORT P346 24 METHODOLOGY gt The minimum guantity of tracer should be injected consistent with the following two requirements gt The maximum concentration at the Tees Intake must be no more than 0 lug l above background This is an absolute requirement of each study Note in the Greta studies there is considerable dilution between the most downstream monitoring point and the intake There should be clear resolution of the concentration curve at each monitoring point as described in the document DEFINITION OF ACCEPTABLE DATA The only acceptable tracer is Rhodamine WT Immediate contact should be made with the water undertaker if these values are likely to be exceeded at the point of abstraction The total length of river specified must be traced at one time for any flow condition However more than one injection may be used if agreed beforehand with the Project Manager Monitoring
10. validated and outputs delivered by 12th February 1999 It will therefore be acceptable if the three surveys are not all in different flow bands Details of individual surveys should be agreed with the project leader in advance The EA can provide assistance and on line river flow information during the working day Each survey should be undertaken in accordance with the acceptable data document appended The survey methodology appended should be adhered to strictly in order to maintain healthy amp wholesome public water supplies 4 0 Project Outputs 4 1 Data Raw data validated data corrected to background concentration spurious data removed etc a data summary and a study summary should be provided in EXCEL format A template 15 provided only mandatory fields need be completed Ownership of the data is the sole right of the EA and can be provided by the EA to third parties at its discretion The data collected for the purpose of this project must not be sold by the contractor used for any other projects or distributed without the consent of the EA Provision should be made by the contractor to provide the data or information about the data at any time throughout the contract in order to fulfil operational requirements e g pollution incidents 4 2 Reports One final report should be provided at the end of the project This should include a comprehensive discussion of the data its value in incident prediction flow conditions of su
11. velocities obtained for the first arrival of the 10 concentration value c 10 are of the order of 10 greater than the mean velocities whilst the velocity from the tail 10 concentration value C 109 are also around 10 less Predictions of the velocity for c 10 using the ADE analysis exhibit less change compared with the mean This would be expected from a technique that has a limited ability for describing skewed distributions Variations of mean velocity with discharge and reach slope are presented in Fig 6 2 and 6 3 respectively A general trend of increasing velocity with discharge is evident in Fig 6 2 however the individual reach results obtained during a single injection appear to mask any significant trend It is also suggested that to allow for the different scales of rivers tested a parameter describing discharge perhaps non dimensionalised with the long term mean discharge may provide a better insight Variation of mean velocity with channel slope Fig 6 3 suggests an increasing velocity with increasing slope again with much scatter It is clear that in it s current form this data 15 of limited use partially due to the method of obtaining slope information but also due to a lack of additional information describing the channel type and boundary material R amp D TECHNICAL REPORT P346 15 Velocity m s P 0 44 COP OO 6 et oa oo elon o 66 oe 4
12. your hard disc in Win95 or NT4 Windows Explorer can be used to do this Running the database The database can be run by either of the two following methods A starting Access 97 and loading the file TOT mde B double clicking on the file s icon in either Windows Explorer or My Computer for Win 05 or NT4 The location of the file TOT mde will depend on vvhereabouts you have copted it to on your hard d sc You may wish to create a shortcut to the database to place on your desktop Win 95 users can do this by clicking the right mouse button over an empty area of desktop and selecting Nevv and then Shortcut The Browse fac l ty will allow you to locate and select the TOT mde file The Main Menu Once the database has loaded successfully the main menu will be shown Depending on your display setup this vvill look similar to Figure 1 belovv Figure 1 The Main Menu Time of Travel Database Please choose vvhat to do by clicking one of the buttons below Search Button mport Remove Quit Button Button Button R amp D TECHNICAL REPORT P346 35 Depending on what you wish to do you may click one of the three main buttons or the Quit button Eziting the Database If you wish to leave the database you should click the Qurt button on the main menu Figure 1 Adding Data to the Database There is only one method provided to add data to the database This is t
13. 5 Initial Results Menu ez x The following tests contain data which matches the criteria you set More detailed output can now be produced Results List Tabular Form Button Select the r m you require search output to take or click Cancel to return to setting criteria Report Form aes Button Button Producing Output From the initial results window you can choose to produce tabular or report form output Report Form Output If report form output 1s chosen a window appears giving the choices for report form output Using the checkbozes you may select which sections to include in your report Using the Report Title box you may enter a title for your report After selecting the required sections and setting a title you can produce the report by clicking the OK button The report will open in Print Preview mode Using the buttons at the top of the screen you may choose to print the report Tabular Form Output If tabular form output is chosen a window appears giving the choices for tabular form output Using the checkboxes you may select which information to include on your table These checkboxes relate to groups of columns for example checking the River Details box will add three columns to the table River Name Area and Region After selecting the required columns you can produce the table by clicking the OK button The table will then be shown You m
14. 68 Wallis S G Guymer I and Bilgi A 1989b A Practical Engineering Approach to Modelling Longitudinal Dispersion Proc of Int Conf on Hydraulic and Environmental Modelling of Coastal Estuarine and River Waters Bradford England 19 21 Sept 291 300 Wallis S G Young P C and Beven K J 1989a Experimental Investigations of the Aggregated Dead Zone Model for Longitudinal Solute Transport in Stream Channels Proc Inst Civil Eng Part 2 87 1 22 Young P Jakeman A amp McMurtrie R 1980 An Instrument Variable Method for Model Order Identification Automatica Vol 16 281 294 Young P C and Wallis S G 1986 The aggregated dead zone ADZ model for dispersion in rivers Int Conf on Wat Qual Modelling in the Inland Nat Env Bournemouth Eng L1 421 433 R amp D TECHNICAL REPORT P346 23 APPENDICES 10 1 DEFINITION OF ACCEPTABLE DATA METHODOLOGY AND SPECIMEN TENDER DOCUMENT The following points define the minimum standard of data reguired to fulfil the objectives of this project 1 The minimum concentration resolution should be 1 40 of the measured peak concentration after background removal Any perceived difficulties in achieving this standard should be resolved with the Project Manager prior to commencing the survey Much greater resolutions are however preferred 2 Tracer arrival and peak should be clearly defined 3 The tail should recede at least to 10 of the
15. A National Database of Travel Time Dispersion and Methodologies for the Protection of River Abstractions Technical Report P346 A National Database of Travel Time Dispersion and Methodologies for the Protection of River Abstractions R amp D Technical Report P346 Dr I Guymer Research Contractor Department of Civil and Structural Engineering University of Sheffield Publishing Organisation Environment Agency Rio House Waterside Drive Aztec West Almondsbury Bristol BS32 4UD Tel 01454 624400 Fax 01454 624409 Website www environment agency gov uk ISBN 1 85705 8216 Environment Agency 2002 All rights reserved No part of this document may be produced stored in a retrteval system or transmitted in any form or by any means electronic mechanical photocopying recording or otherwise without the prior permission of the Environment Agency The views expressed in this document are not necessarily those of the Environment Agency Its officers servants or agents accept no liability whatsoever for any loss or damage arising from the interpretation or use of the information or reliance upon the views contained herein Dissemination status Internal Released to Regions External Released to Public Domain Statement of Use This report is intended to provide an increased understanding of the fundamental river travel time and dispersion processes The report describes work undertaken to develop a searchable database fo
16. HNICAL REPORT P346 1 e Totest new predictive technigues against pilot catchment data sets 1 3 Key Tasks for the Project Collection of data Database design and structure Ouality control Validation of data Calculation of parameters coefficients from data Derivation of predictive methods empirical stochastic from pilot data CD ROM output incorporating user friendly database and predictive equations R amp D TECHNICAL REPORT P346 2 2 SUMMARY THEORY Soluble pollutant instantaneously released into a river may be considered to be subjected to two processes advection and dispersion In practise these processes are linked with several physical mechanisms contributing to the overall observed effects However the problem 1s simplified if it is conceptualised as advection causing a bodily transport of the pollutant downstream and dispersion causing a spreading This spreading leads to an associated reduction in the peak concentration Soluble dye tracers have been used for several years to quantify these effects in rivers and an example of good quality data collected from such a study is reproduced in Fig 2 1 A description of acceptable data is provided in Appendix 10 1 Rutherford 1994 provides an excellent practical guide to river mixing 0 9 gt 0 8 0 7 m Nn 500 1000 1500 2000 2500 Time Concentration A o O w N Fig 2 1 Example of Temporal Concentration Distributions Recorded at Se
17. I A mn Naw 2 0 8 x 0 95 0 7 1 A R 0 6 5 p NYY RTT T TULU zal RE O 0 2 IZI 2 4576 IZI 0 91 12 7 2 0730 e 0 1 p 1 2288 2 v 1 7658 0 6144 aS d RES K 0 9 1 4586 es 295 2 97 299 3 01 3 03 3 05 9 0 00 1 20 2 40 3 60 4 80 6 00 Travel Time Travel Time Fig 4 2a Fig 4 2b Fig 4 2 Surface Representation of Matriz Optimisation Results On completion of the optimisation procedure for each reach all the standard and optimised coefficients were given along with the R values for each case A hard copy summary sheet was also produced and these have been collated in the Project Record Appendix 6 0 R amp D TECHNICAL REPORT P346 12 PRODUCTION OF ACCESS DATABASE A database has been designed and developed to hold information pertinent to reaches that have been studied and whose data were considered acceptable see 3 0 Data collation If a reach has been studied and the data accepted more than once then there will be multiple entries for that reach The database was produced by Mr Tom Tunstill as part requirement for the MSc in Information Systems supervised by Dr Ian Guymer Below is reproduced the abstract from the thesis which is provided in full in the Project Record Appendix 5 0 A User Manual is reproduced in Appendix 10 3 the development of a database
18. al Mean Flow m3 s 143 Theoretical Q95 Flow m3 s J43 Logging Started Time hh mm K43 Date dd mm yyyy L43 Time Since Injection minutes M43 Additional Comments G35 amp G36 Chart Format Dye Conc Y Azis Unit PPB ug l etc B26 Time Plot X Axis Unit Days Hours Minutes or Seconds B27 Data Logging Method B28 Data Logging Interval Seconds B29 Date Logging Started dd mm yyyy amp hh mm B30 amp D30 Date Logging Ended dd mm yyyy amp hh mm B31 amp D31 Raw Data Sheet For each monitoring station the name is stored in cell B and the data listed below in columns A and B from rows 11 onwards Subsequent data is stored in the same format translated 4 columns to the right 1 e for the monitoring station name columns E H K etc see attached example print out R amp D TECHNICAL REPORT P346 30 pp wurqy uon lu 1912 amy aun payers Burbbo7 15 cul 801 660 mzuzu shieqsiq 1230510341 1230310341 OLOPE OLOPE 28 891 WI T qumn yJ aua 709 HO 9 t200 2250 89 0 921 920 sgu 0 uea 4 3 3 SHU UWO sabneg zods jo on s w uoneys Buibneg 02 mold G60 sew moj Al pq abesaay s w 44014 uean Ajeg s u uon lu yo aun abseyssiq zwy uoneys Huilbnes 104 ease yuaulu 5
19. ation plot of all the data recorded at each station and for each of n measurement stations Sheets 3 1 to 3 n A plot of the individual concentration time distribution Station name In the interests of storage the individual plots were later removed Examples of the agreed format are provided in the Appendix 10 2 It was also agreed that all future data to be included within the database would only be accepted in this format R amp D TECHNICAL REPORT P346 9 4 ANALYSIS OF DATA To remove the variability in the prediction of travel time and dispersion parameters due to data quality and accuracy a method of optimising predictions of a downstream temporal concentration distribution from an upstream distribution was developed Firstly it was necessary to mass balance the data This procedure assumed that the tracer was conservative the river flow constant and that the entire tracer mass measured upstream was also measured at the downstream site To achieve a mass balance the downstream data points were each multiplied by a mass balance factor This factor was calculated from the upstream tracer mass divided by the downstream tracer mass A FORTRAN program was written which could calculate the standard ADE and ADZ coefficients from a given pair of temporal concentration distributions These are the coefficients used in equations 2 1 and 2 3 for predicting downstream temporal concentration distributions For each temporal concentration d
20. ay now print the table or copy and paste information from it into other applications e g MS Excel R amp D TECHNICAL REPORT P346 41
21. ble for the dispersion of the tracer As such its variation may be considered comparable to the longitudinal dispersion coefficient used in the ADE technique R amp D TECHNICAL REPORT P346 7 3 DATA COLLATION Several attempts between the start of the project in November 1997 and April 1999 were made to obtain electronic copies of river travel time data for inclusion in the analysis Table 3 1 provides summary details by Region Considerable resources were employed to collate this data and convert it into a consistent format Once the data had been converted to standard Microsoft Excel format and graphical hard copies produced these copies were used to make the initial decision whether to include the data for the analysis Agency 1 Call 27 Call Final Call Optimised Data Region deadline deadline Accepted Data included in 12 Nov 98 1 April 99 Database Midlands Table 3 1 Details of Data Obtained and Included for Analysis 3 1 Acceptance of Data A preliminary subjective decision whether to include data within the first round of analysis was made based on the visual inspection of the distributions It had been agreed that tracing events would only be included if multiple stations were used for measurement This is due to the problems of uncertainty in injection method 1f only one downstream measurement station was used All the tracing data provided by Thames Region comprised a single measurement station and has therefore not been i
22. charge Trends 6 3 Longitudinal Dispersion Trends 7 CONCLUSIONS 8 RECOMMENDATIONS REFERENCES R amp D TECHNICAL REPORT P346 1 ill iii IV 10 13 14 14 18 21 22 23 APPENDICES 10 1 Definition Of Acceptable Data Methodology and Specimen Tender Document 24 10 2 Format of Excel Data Store 29 10 3 Access Database User Manual 35 within Project Record PR 346 1 Appendix 1 0 Membership of the National Sub Committee Appendix 2 0 List of Meetings Appendix 3 0 Minutes of Meetings Appendix 4 0 Scoping Report Appendix 5 0 Development of Access Database Appendix 6 0 Hardcopy of Excel Data Store R amp D TECHNICAL REPORT P346 ii LIST OF FIGURES 2 1 Example of Temporal Concentration Distributions Recorded at Several Sites 22 Illustration of ADE Technique z Representation of an ADZ prediction for a single unit 2 4 prediction showing both the advection and decay elements Illustration of ADZ Parameters 3 1 Examples of Poor Quality Data 4 1 Representation of the matrix optimisation procedure 4 2 Surface representation of matrix optimisation results 6 1 Histogram of Reach Mean Velocities 6 2 Variation of Mean Velocity with Discharge 6 3 Variation of Mean Velocity with Reach Slope 6 4 Comparison of Predicted Peak Velocities Using Jobson 1997 Technique 6 5 Results for Reaches with Five Traces Undertaken 6 6 Results for Reaches with Four Traces Undertaken 6 7 Variation of Dispersion Coeffici
23. d be assumed that mains power will not be available Communication should be established between the contractor and EA for the duration of the surveys Results should be made immediately available to identify any immediate risk to intakes As far as it 1s possible to pre plan tracing studies should not be undertaken in adverse weather conditions Tracing studies should be undertaken at constant discharge for the duration It is accepted that this may not be possible to strictly adhere to for high flows The EA will provide one liaison person for each experiment All EA Health amp Safety requirements as described in the Specification should be fully adhered to R amp D TECHNICAL REPORT P346 25 Time of Travel amp Dispersion Studies River Tees Risk Assessment 1 0 Proyect Outline Aim To collect and validate time of travel data for the River Greta and to produce pollution incident tables Objectives 1 To collect time of travel and solute dispersion data across three flow bands for the river Greta 2 To validate this data and derive pollution incident tables appropriate for use by field staff 2 0 Background It is the intention of the North East Region of the Environment Agency EA that each of its riverine potable abstractions should be protected from pollution incidents This will be achieved by the collection of appropriate field data which will be used to develop predictive techniques describing the advection and dis
24. dinal dispersion of solutes in small river systems Young amp Wallis 1986 Barraclough et al 1994 This model assumes that a tracer is advected through an entire reach by plug flow 1 e advection with no dispersion after which it passes through a single mixing cell that has the aggregated effect of all the dead zones within the reach 1 e dispersion with no advection These processes are illustrated for a single concentration unit and for a complete temporal concentration distribution in Fig 2 3 and 2 4 respectively The term dead zone is often misunderstood although it implies a form of pocket that is separated from the main flow it should be considered in a wider context as a bulk parameter that not only describes the effect of segregated regions of flow but also other dispersive catalysts such as eddies viscose sub layers and velocity profile from Wallis Young amp Beven 1989a 6 Input T 5 o O C O O Advection Exponential decay Time Fig 2 3 Representation of an ADZ prediction for a single unit R amp D TECHNICAL REPORT P346 5 6 Upstream measured concentration o O O 10 Dovvnstream dq b predicted 57575 concentration 1177 Advection Time Fig 2 4 ADZ prediction showing both the advection and decay elements In practice data is often acquired at discrete sampling times rather than as a continuous time varying fluctuation Within Fig 2 5 the da
25. e operator would be is greater than and the specified value would be 0 95 Removing Criteria After setting criteria you may wish to remove one To do this select the criteria you wish to remove from the criteria list Figure 4 by clicking on it The selected criteria will be highlighted To remove it click the Remove button Figure 4 Editing Criteria After setting criteria you may wish to edit it To do this double click on it in the Criteria List Figure 4 The values you set for it will appear in the Attribute Setter Operator setter and Specified Value box You can the edit these After editing you can update the criteria by clicking the Add Update button Figure 4 The updated criteria will be shown in the Criteria List Figure 4 Search Menu Attribute Criteria Type Operator Setter Setter Setter 22 x Use the c vntrols below to set a output criteria then clic Add Update to add it to the list of citeria shown Edit a criteria doublec ek it in the list below Specified Value Box Add Update Button To Remove a criteria select it by clicking on it in the list above then Remove click Remove Button After Setting Criteria click Search to search the database or Cancel Search to return to the main menu Button R cD TECHNICAL REPORT P346 39 Searching the Database After you have set your search criteria y
26. e required for the solute plume to pass a point in the river Data is summarised from different rivers representing a wide range of sites slopes and geomorphic types Jobson 1997 collated data from 90 US rivers and obtained a data set where most variables were available for 939 reaches Four variables were available in sufficient quantities to undertake a regression analysis These variables were the drainage area D the reach slope S the mean annual discharge Q and the discharge at the time of measurement Q To study their relationship with the velocity of the peak concentration vp it was reasoned that the variables could be combined into two dimensionless groups the dimensionless drainage area defined as D 0129 Q where g is acceleration of gravity and the dimensionless relative discharge defined as Q Q A linear relationship was assumed between the peak velocity and the product of three terms the dimensionless drainage area D raised to a power x the dimensionless relative discharge raised to a power y and Q D The relationship was optimised to obtain the most accurate fit to the available data by varying four parameters the intercept and slope of the assumed linear relationship and the powers x and y and produced this an equation with an R of 0 7 Jobson acknowledges that the travel time is the least accurately defined relationship and although the data available to this current Agency study is less compr
27. e results collected from individual river time of travel studies commissioned by separate Regions within the EA The aim was to realise the potential to significantly increase current understanding of the fundamental river travel time and dispersion processes A review of recent work related to estimating travel times and dispersion characteristics on untraced rivers is presented in Green et al 1994 However without resorting to complex numerical techniques to estimate uncertainty it is felt that combining and interpreting results obtained on other similar watercourses could lead to an improved estimate of the travel time and longitudinal dispersion This has been performed in the U S by Jobson 1997 The following are reproduced from the original contract 1 1 Overall Project Objective To provide each region with a national time of travel and dispersion database coupled with simple empirical equations for predicting the travel time and spread of pollutant in a river catchment 1 2 Specific Objectives To summarise current thinking theory and practise within the Agency and externally To visit all eight regions to collect copies of existing data To analyse all existing data for key travel time and dispersion parameters To build a database to hold the raw data summary data and equations To develop new predictive empirical techniques for untraced ungauged river reaches and associated uncertainty bands or catchment datasets R amp D TEC
28. ehensive similar relationships have been developed and results are illustrated in Section 6 0 2 1 Advection Dispersion Model ADE The results of the work by Taylor 1954 predict instantaneous spatial concentration profiles of Gaussian shape Often in practise a simple routing procedure is used to predict the temporal concentration distribution c x t at a downstream site x knowing the temporal concentration at an upstream site x Fischer et al 1979 This is summarised by za a 2 x gt t ap 2 1 4nD t gt t 4D t gt ty where D the longitudinal coefficient y an integration variable ti the time of passage of the centroid centre of mass of the tracer cloud at site 1 given by a 2 2 This method effectively takes each individual upstream element of the temporal concentration distribution advects it downstream by a fixed amount and spreads it assuming a Gaussian distribution The individual Gaussian distributions are then summed at each time to obtain the overall downstream temporal concentration distribution This is illustrated in Fig 2 2 below R amp D TECHNICAL REPORT P346 4 Travel Time t Concentration Centre of Mass Fig 2 2 Illustration of ADE Technique 2 2 Aggregated Dead Zone Model ADZ The aggregated dead zone ADZ model Beer amp Young 1984 Wallis et al 1989a has gained favour by practitioners in the UK wishing to describe the longitu
29. ent with Discharge 6 8 Variation of Dispersive Fraction with Discharge 6 9 Variation of Dispersion Coefficient with Dispersive Fraction LIST OF TABLES 3 1 Details of Data Obtained and Included for Analysis 4 1 Ultimate matrix resolution Values 6 1 Summary of Reach Mean Velocities 6 2 Parameters Derived from Fitting v kQ Trends R amp D TECHNICAL REPORT P346 111 QI Q d U 12 14 16 16 17 17 18 19 19 20 12 15 18 EXECUTIVE SUMMARY A knowledge of solute travel times within rivers is important for the protection of potable water supplies and in the development and calibration of river and catchment water quality models This report describes work undertaken to obtain added benefit by incorporating all recent tracer study results into a single searchable database Both practical and theoretical approaches for describing advection and dispersion have been described Results from solute tracer studies where the raw data were available in electronic form have been collected from each region of the Agency These have been plotted in a consistent format A standard data storage format has been developed and the acceptable data sets have been formatted This format includes all details of the river traced region catchment area the contractor that undertook the work the flow conditions both at the time of the trace and historical the tracer used and the reaches studied The datastore containing over one hundred differen
30. etc or any combination of these The database can also be searched for reaches or rivers which have had specific numbers of tests performed on them Setting Criteria To begin setting criteria click the button labelled Search for Output data on the main menu Figure 1 After a short pause another window should appear This is the Search Menu It should look something like Figure 4 The best way of explaining the criteria setting process is to give an example Criteria Setting Example Say that you wanted to search for test data for the Midland region where the Rt value for the ADE analysis was greater than 0 95 You would need to set the following criteria 1 Region is Midland 2 ADE Rt2 ss greater than 0 95 To set the first of these the following steps would be taken 1 Using the criteria type setter Figure 4 choose Geographical from the list since this is a geographical criteria 2 Using the attribute setter Figure 4 choose Region from the list 3 Using the operator setter choose is from the list 4 Type Midland into the Specified value box Figure 4 R amp D TECHNICAL REPORT P346 38 5 Add this criteria by clicking the Add Update button Figure 4 6 The criteria will be displayed in the Criteria List Figure 4 Similar steps would be taken to add the second criteria only this time the criteria type would be Analysed Results th
31. exact fit to the measured downstream data would give a value of unity for R amp D TECHNICAL REPORT P346 10 R and a value of less than zero would indicate that the prediction fails to describe any part of the measured result Having produced an initial prediction using the determined standard parameters in equations 2 1 and 2 3 the values of the parameters were optimised to improve the prediction For the ADE and ADZ equations the optimisation procedure was very similar A sequence of refined searches through combinations of parameters travel time and dispersion coefficient values for the ADE equation and travel time and time delay for the ADZ model were performed to determine the pair which gave the best fit to the downstream temporal concentration profile A matrix system was employed which greatly reduced the number of calculations required to reach the best fit solution The pair of coefficients which gave the prediction with the best fit to the downstream data represented by the cell with the greatest R value assigned to it were determined A new matrix was created by the programme which zoomed in towards the best fit coefficients This procedure is shown diagrammatically in Fig 4 1 The process of producing a new matrix was repeated until a predetermined final resolution was obtained The final resolutions used are given in Table 4 1 Coefficient A 0 5 0 6 0 7 0 8 0 Coefficient A
32. g the effect of discharge changes on the travel time was undertaken using equations of the form v kQ where v is the mean velocity the discharge k a constant and x the power In collating the results from this study values of k and x have been obtained for reaches where more than three travel time studies have been performed These are shown in Fig 6 5 and 6 6 for reaches with five and four results respectively However the reaches that have five repeat traces performed have two repeat tests undertaken at similar discharges and are of little additional benefit than those with four traces Values for the fitted parameters are presented in Table 6 2 0 6 0 5 Velocity m s c aN N 0 1 0 2 4 6 8 10 12 14 16 18 20 Discharge m3 s Fig 6 5 Results for Reaches with Five Traces Undertaken R amp D TECHNICAL REPORT P346 17 0 6 gt 0 5 1 0 4 gt 0 3 Velocity m s 0 2 0 1 gt 0 5 10 15 20 25 30 35 40 Discharge m3 s Fig 6 6 Results for Reaches with Four Traces Undertaken It is difficult from such a small data set to obtain clear trends with any degree of confidence Values of the power range from 0 6100 to 0 9777 with the constant ranging between 0 01 and 0 1078 It appears that the power increases with decreasing river slope however further results would be required over a wider range of slopes with more discharges to provide any definite relationship
33. information may be obtained by expanding the database in this way This is strongly recommended The paucity of available data has severely limited the level of analysis and applicability of predictions Improved accuracy of reach lengths alone would improve accuracy of velocity and dispersion parameters Inclusion of bed levels at all monitoring sites would allow the determination of reach slopes for all the available data This would significantly expand the available data set It has not been possible to test the predictive techniques against catchment data sets At this stage it 1s suggested that further development of the database is undertaken More specifically it is possible that greater confidence can be gained by looking at the goodness of fit of the analysed results This should be followed by an attempt to obtain additional information regarding the river reaches that have already been traced Information such as an accurate measurement of bed slope channel size and cross sectional shape if possible as a function of longitudinal distance Planform curvature is also a channel property that will significantly influence longitudinal dispersion and one that may be readily obtainable from GIS systems This should reduce some of the scatter observed on the relationships and increase the confidence of predictions This would allow the working envelope of the Jobson 1997 analysis to be determined Further it may be that additional useful info
34. is file If part of this data is shared with other tests still in the database this part will not be removed To remove data from the database use the following steps 1 Click the button labelled Remove Data Relating to a Test on the main menu Figure 1 After a short pause another window should appear This is the Delete Menu It should look something like Figure 3 2 Using the File Selector Figure 3 select or type the filename for the test which you wish to remove from the database 3 Click the OK button to begin removing data If the OK button 15 not enabled please check that you have selected a filename correctly 4 A message will be displayed showing that you are deleting data 5 When the removal process has finished a message will be displayed stating this You may now click the OK button to return to the main menu R amp D TECHNICAL REPORT P346 37 6 Figure 3 The Delete Menu x Select or type the filename relating to the test data you wish to delete using the control below Then click OK to delete the data from the database aaa File Selector 614 Cancel Button OK Button Searching for data and producing output The database can be searched using search criteria These may be geographical criteria e g a specific river name or injection point physical criteria e g reach length or slope or Analysed results criteria e g Rt values ADE Velocities
35. istribution the time of the first arrival of the tracer was noted and the value for the centroid time equation 2 2 was determined From these values both the mean velocity of the tracer cloud given by X2 uU 4 1 7 dt 441 and the temporal variance o X at site i given by oo t tiy eG that of xi 4 2 ci dt oo were determined A value for the longitudinal dispersion coefficient D was then determined rom pave eten 4 3 2 t2 t1 These values for centroid travel time t dispersion coefficient D and reach time delay r were obtained directly from analysis of the raw data and as such have been termed standard coeffictents They are strongly influenced by the quality of the raw data and decisions on both the start and end times for the temporal concentration distributions An optimisation procedure was developed to minimise the influences The ADE and ADZ equations 12 11 and 2 3 were then used in combination with these standard coefficients to predict downstream concentration profiles from the upstream data A measure of the goodness of fit between the profiles and the actual data R Young Jakeman amp McMurtrie 1980 was calculated The determination of R is given by equation n S pi 2 t 1 Ri 1 1 14 4 20 t 1 where c and p are the measured and predicted data values at time t Using this definition a prediction with an
36. ls of the tracing exercise together with a plot and listing of the raw data These data have been analysed for advection and dispersion characteristics and the results for each reach appended to the Excel files A MS Access database has been developed that allows complex search criteria to be adopted This database has been used to define the variation of solute velocity within the reaches studied and also to develop trends relating the travel velocity to the discharge To a lesser extent relationships with river bed slope have also been obtained This has produced useful results but is limited by the quality and quantity of available data Descriptions of current theories and approaches for describing and predicting advection and dispersion have been summarised Guidelines have been developed to assist with collecting appropriate and accurate information in future time of travel studies R amp D TECHNICAL REPORT P346 21 8 RECOMMENDATIONS Considerable resources have been employed both within and external to the Agency to produce this database and it is essential that ownership is taken and it is maintained thoroughly It has the capacity to be extended to include new surveys and it s usefulness will then be enhanced The database has been structured to allow for the inclusion of historical data without the need to undertake optimisation analysis Although the confidence level associated with this data may be reduced significant additional
37. n where n is excedence B6 Reach Name if applicable B7 Contractor Details Project Reference J2 Contracting Company J3 Company Contact J4 Telephone Number J5 Address J6 amp J7 Injection Information River trib watercourse name B11 Dye Injected Time hh mm B12 Date dd mm yyyy D12 at Location B13 Bed level m A O D B14 Tracer B15 Type of tracer used Dosing Volume Litres B16 Dosing Concentration mg l B17 Catchment Area B18 upstream of injection point Grid Reference B19 Theoretical mean flow m3 s B20 Theoretical Q95 Flow m3 s B21 Hydrological reference B22 Flow Data Ref Gauging Station J13 Grid Ref K13 Hydrological Reference L13 EA Number M13 Catchment area for Gauging station m3 s J14 Discharge Time of Injection m3 s J15 R amp D TECHNICAL REPORT P346 29 Daily Mean Flow m3 s J16 During test Average Daily Flow m3 s J17 Over long period Q95 Flow Gauging station m3 s J18 No of Spot Gauges J19 If undertaken Comments J20 amp J21 Summary of Spot Gauges rows 28 to 31 Number G28 Grid Ref H28 Time hh mm 128 Discharge m3 s J28 Theoretical Mean Flow m3 s K28 Theoretical 095 Flow m3 s L28 Hydrological reference M28 List of Monitoring Points rovvs 43 upvvards Number A43 Names B43 Grid Ref C43 Elevation m A O D D43 River Trib Watercourse Name E43 Catchment Area km2 F43 Hydrological Reference G43 River Distance from Injection Point km H43 Theoretic
38. ncluded within this first round of analysis The main factors that have been used to decide on inclusion of the data are the completeness of the distributions do they exhibit a clearly defined start peak and reduction of tracer concentration towards background as defined in section 10 1 and data recording interval this needs to be constant and equal at both the up and downstream sites on a reach An example of good data that has been included for analysis is shown in Fig 2 1 whilst examples of poor quality data illustrating the main reasons for discarding data are shown in Fig 3 1 For example the noise exhibited in Fig 3 1b precludes any useful information on either travel times or dispersion 3 2 Development of Spreadsheet Data Format It was agreed that data would be stored in two forms Firstly a file Microsoft Excel was to be created to contain all background information regarding the actual river tracing exercise including details of the contractor hydrological conditions and all the raw data minutes of meeting 24 Feb 98 Secondly a database Microsoft Access was to be constructed to contain the summary results after analysis of the raw data R amp D TECHNICAL REPORT P346 8 0 20 5 0 002 048 4 0 0018 4 0 16 4 0 0016 4 5 0 0014 1 2 0 14 4 wa 0 0012 bl 0 12 4 A bi b s 5 it bo 0 001 4 ni bo 5 0 10 E 1 m 8 1 8 5 0 0008 5257 0 08 4 0 0006 He Po SE 0
39. nd Safety requirements change R amp D TECHNICAL REPORT P346 iv L INTRODUCTION The protection of potable intakes from pollution is of paramount importance to ultimately maintain the integrity of public health It is a statutory duty of the Environment Agency to protect private water abstractions Often there are multiple abstractions on one river and it is the Agency s duty to protect these from pollution incidents Over the past 10 years in particular a great number of dye tracing studies have been completed through the Agency and formerly the National Rivers Authority NRA These studies have provided valuable details of travel time flow relationships and the dispersive characteristics of river reaches In turn this data has been successfully used to predict travel times of major pollution spillages such as the Asulam incident on the river Swale in Yorkshire in 1994 However dye tracing is expensive Usually tracer studies have been designed to provide data for the most critical part of the catchment e g 30 kilometres upstream of the major abstractions To build a sensible scientific model to predict the passage of a pollution spillage a minimum of three independent tracer studies repeated for the same river reach are required It is therefore impracticable to collect a complete data set for every drinking water catchment in England and Wales The costs would be prohibitive This project was awarded to collate validate and summaris
40. o import data from an Excel file The Excel file must be in the standard format paying particular attention to the cell locations Included with this database should be a set of Excel files from which data has already been imported If you are unsure of what the standard format is please refer to these To import data from an Excel file use the following steps 1 Click the button labelled Import Data from Excel Sheet on the main menu Figure 1 After a short pause another window should appear This is the Import Menu It should look something like Figure 2 2 First select the drive from which to import the file using the Drive Selector see Figure 2 3 You then need to select the path where the file to be imported is stored The current path is shown in the Current Path Display Figure 2 You can use the Path Selector to navigate to the path you require by double clicking on a path s name to set it as the current path Once you are satisfied that you have selected the correct path you are ready to select the file 4 You may select the file you wish to import data from using the File Selector Figure 2 If you have selected the path correctly in the previous step the filename you require will be shown as one of the list in File Selector To select it double click on it The selected path and file will be shown in the Selected path and file displays Figure 2 Ensure that Excel is no
41. ou may now use them to search the database If you have set no criteria all data in the database will be given as output To search click the Search button on the Search Menu Figure 4 After a short pause another window should appear This is the Initial Results Menu t should look something like Figure 5 Shown in this window is the Initial Results List Figure 5 This is a summary of which tests contain data that matches the criteria set for the search and is intended to give some idea of the number of results found Initial Results At this stage you can return to setting or editing criteria by clicking the Cancel button If however you want to produce output from the search you may choose one of two alternatives These are described below Report Form Produces output intended for printing out as hard copy This output is arranged in a similar way to the standard Excel file The sections to be shown on the report can be chosen This is useful to check what data is stored in the database To produce Report Form Output click the Report Form Button Figure 5 Tabular Form Produces a table of results The data that is included in the columns can be chosen This is particularly useful when examining possible relationships between pieces of numerical data To produce Tabular Form Output click the Tabular Form Button Figure 5 R amp D TECHNICAL REPORT P346 40 Figure
42. persion of pollutants These measures will in many cases be supported by catchment based Risk Assessment Studies which will locate categorise and where possible minimise the quantities of pollutants in a catchment and develop strategies for dealing with pollution incidents One such Risk Assessment Study is ongoing in the Tees catchment A significant risk already identified is that of road tanker accidents on the A66 T This project aims to provide the necessary data and capability to effectively manage such accidents which result in pollution of the River Greta and which pose a threat to the Tees Intake 3 0 Tracer Surveys Surveys to be undertaken Approximate length to be traced Number of surveys at flow bands Km Low Medium High The flow bands are defined as follows Low lt 20 ile Medium 20 55 ile High 55 ile R amp D TECHNICAL REPORT P346 26 Survey sites Injection Railway Bridge NY 900 119 Monitor 1 1 Bowes Bridge NY 995 133 Monitor 2 Rutherford Bridge GS NZ 034 122 Monitor 3 1 Rokeby Park Footbridge or NZ 087 139 Dairy Bridge NZ 084 144 Note A decision on which of the two site should be selected for Monitor 3 must be agreed with the EA At Rokeby Park Footbridge permission will have to be obtained from the owners of the estate The above flow bands are preferred It 1s required however that the project 1s completed this financial year 1998 99 All three studies should therefore be completed data
43. r use by EA staff particularly those involved in contracting out tracer studies or required to give predictions in the event of an incident The report should also be useful to water quality staff involved in river modelling Keywords river tracing pollution rivers solutes advection dispersion travel times database Research Contractor This document was produced under R amp D Project P346 by Dr Ian Guymer Department of Civil and Structural Engineering The University of Sheffield Sir Frederick Mappin Building Mappin Street Sheffield S1 3JD Tel 0114 222 5353 Far 0114 222 5700 Environment Agency Project Manager The Environment A gency s Project Manager for R zD Project P2 092 was Mr T Hardy North East Region Further copies of this report are available from Environment Agency R amp D Dissemination Centre WRc Frankland Road Swindon Wilts 5 5 8YF Tel 01793 865000 Fax 01793 514562 E mail publications wrcplc co uk R amp D TECHNICAL REPORT P346 CONTENTS LIST OF FIGURES LIST OF TABLES EKECUTIVE SUMMARY 1 INTRODUCTION 1 1 Overall Project Objective 1 2 Specific Objectives 1 3 Key Tasks for the Project 2 SUMMARY THEORY 2 1 Advection Dispersion Model ADE 2 2 Aggregated Dead Zone Model ADZ 3 DATA COLLATION 3 1 Acceptance of Data 3 2 Development of Spreadsheet Data Format 4 ANALYSIS OF DATA a PRODUCTION OF ACCESS DATABASE 6 PREDICTIONS TRENDS 6 1 Travel Times 6 2 Velocity Dis
44. rmation is available within the Environment Agency River Habitat Survey database and a merger of the information may make a significant contribution to predicting travel times and dispersion within catchment modelling processes R amp D TECHNICAL REPORT P346 22 REFERENCES Barraclough A Freestone R Guymer 1 and O Brien R T 1994 Evaluation of the Aggregated Dead Zone ADZ method as a River Catchment Management Tool applied to the rivers Aire and Derwent in Yorkshire 2nd Int Conf on Hyd Modelling Stratford upon Avon U K 14 16 June 439 449 Beer T amp Young P C 1984 Longitudinal Dispersion in Natural Streams Proc A S C E J Env Eng Div 109 1049 1067 Fischer H B List E J Koh R C Y Imberger J and Brooks N H 1979 Mixing in Inland and Coastal Waters Academic Press New York Green H M Beven K J Buckley K amp Young P C 1994 Pollution Incident Prediction with Uncertainty Mixing and Transport in the Environment K J Beven P C Chatwin amp J H Millbank eds John Wiley amp Sons 113 137 Jobson H E 1997 Predicting Travel time and Dispersion in Rivers and Streams Journal of Hydraulic Engineering Vol 123 No 11 Nov 971 978 Rutherford J C 1994 River Mixing J Wiley amp Sons Chichester England Taylor G I 1954 The Dispersion of Matter in Turbulent Flow Through a Pipe Proc R Soc London Ser A 223 446 4
45. rveys including relevant hydrographs provided by the EA and recommendations R amp D TECHNICAL REPORT P346 21 5 0 Review The EA reserves the right to redesign the tracer surveys within the overall cost boundaries of the contract 6 0 Health Safety The EA will undertake assessments of study sites to enable it to highlight health and safety risk over which it has control The EA will then provide appropriate details to the successful contractor to enable them to carry out their own health and safety risk assessments effectively The successful contractor will be required to work in accordance with the Codes of Practice and Safe Systems of Work sections of the EA Health and Safety Manual where appropriate The successful contractor will however need to demonstrate their awareness of and competency to manage risks associated with any equipment and procedures that may be used in the surveys R amp D TECHNICAL REPORT P346 28 10 2 Format of Excel Data Store The Excel data store comprises a summary sheet a chart showing the data and a vvorksheet containing all the original raw data as provided by the contractor After analysis additional worksheets will be created to contain the results for each reach Summary Sheet Information Cell Location Meaning General Information Version Last Update dd mm yyyy DI File Name B2 Excel file name Region B3 EA Region Area B4 within Region Catchment Name B5 Flow Excedence Category Q
46. shed line represents the actual concentration time distribution of a tracer at two locations up and downstream within a reach Superimposed over these temporal concentration distributions are bars that illustrate the discrete time representation of the data set It is a discretised data set of this form that is used when applying the aggregated dead zone model lt Travel Time t i Concentration C J time step At h Time delay r Time t First First Arrival Arrival Upstream Downstream Fig 2 5 Illustration of ADZ Parameters Wallis et al 1989b give a simple discrete time equation for predicting the temporal concentration distribution at a downstream site for a single cell c x2 t oc x t 1 1 a o x t 6 2 3 where c x t concentration at longitudinal position x at time t with 1 1 or 2 representing the upstream and downstream locations respectively At m residence time t r travel time t t4 i 4 Q time delay t t4 a R amp D TECHNICAL REPORT P346 6 first arrival time at location 1 the discrete time equivalent of the time delay r that is the nearest integer value of 17 4t and At the time step or sampling interval Young and Wallis 1986 define the dispersive fraction as T t D SS 2 4 77 2 4 This parameter quantifies the proportion of the river reach that is assumed responsi
47. should be at a suitably fine temporal resolution to give an accurate definition of the passage of the Rhodamine plume through a site All tracer injections must be of a gulp nature Where not already defined the location of the first sampling point must be sufficiently downstream to achieve initial mixing All reasonable efforts should be made to ensure the plume will not be visible during daylight hours The EA Public Relations Department and Water Company operations must be notified of impending survey work All press releases concerning this project should be handled by the EA The contractor must therefore specify injection times and dates for any surveys at least 24 hours in advance Water samples should be taken from the mainstream section of the channel avoiding any dead zones The same sites identified in the specification should be used in each survey Tenders will be favoured that can demonstrate the use of three fluorometers in series operating at the same time if this is likely to be required given the spacing of the monitoring sites A backup fluorometer should be on hand at all times to ensure gt Continued sampling gt Full protection of water company intakes Tenders will be favoured that can demonstrate the ability to meet the resolution requirements described in DEFINITION OF ACCEPTABLE DATA A reliable source of power should be demonstrated with an acceptable noise level and heavy duty batteries It shoul
48. system intended to store data from time of travel studies carried out on rivers throughout the UK In order to store this data a relevant relational data structure was created Several key features also had to be designed to meet the user s requirements These included 1 an automatic import facility which allows the user to import data directly from an excel spreadsheet file 11 a feature to allow the user to search the database using complex criteria 111 the system should be user friendly This was achieved by designing and implementing a graphical user interface The system has been tested for durability and usability and relevant changes incorporated before production of a final system R amp D TECHNICAL REPORT P346 13 6 PREDICTIONS AND TRENDS Having populated the database with analysed results the final objective of the work was to develop empirical trends relating to both the travel time and the dispersion of material for untraced ungauged rivers In total some 193 reach results were available providing information from 96 different reaches Channel slope was only available for a total of 74 reaches This was provided from OS maps the accuracy of which is illustrated as 3 of the reaches were reported to have negative uphill slopes The paucity of additional data describing the channel properties such as slope channel width planform curvature or bed material restricted the degree to which the development of trends could
49. t running 6 To begin importing data from the file click the OK button Figure 2 If the button is not enabled you have not selected a file correctly please check the previous steps 7 At any time during the previous steps you may click the Cancel button Figure 2 to return to the main menu N Once the import is started step 5 above a message will be shown This message will keep you informed as to what the system is doing The system will first check that the data in the Excel file is valid for input If it is not then a list of problems will be displayed If the import is successful the message Finished importing data will be displayed You may now click OK to return to the main menu R amp D TECHNICAL REPORT P346 36 Figure 2 The Import Menu EZ Use the Controls below to select a file then click to import data from it into the Database 0 7 Drive Selector Current Path Display PParent Directory Path Selector buciq xis Pariner els File Selector OK Button Cancel Button Selected Path and File Display Removing Data from the Database Occasionally you may wish to remove data from the database e g to replace it with updated data The database provides a facility for this To remove data you will need to know the Excel file name s from which it was imported The delete facility will remove data related to the test described in th
50. t tracing studies has been analysed to obtain reliable estimates of the mean travel velocity and the longitudinal dispersion Two techniques have been used for this study an advection dispersion and an aggregated dead zone technique Descriptions of how to calculate advection and dispersion parameters have been included The optimised results from the analysis of each reach have been appended to the spreadsheet datastore A searchable database has been developed and all the results from the data analysis incorporated This software provides the facility to import additional results and to delete and modify the data It allows for complex searches of the database using geographical physical and multiple study criteria The database has been used to obtain quantifiable trends relating the solute tracer velocity and dispersion to available parameters such as discharge and slope The results exhibit significant scatter but relationships have been quantified Recommendations for the maintenance and further development of the database are also included For use in future time of travel surveys the following have been developed minimum methodology for fieldwork definition of acceptable data requirement of summary data to be collected spreadsheet format for holding raw and summary data Also a survey specification template has been included This was satisfactory at the time of publication but may need updating should for example Health a
51. uoness buibneg yay 18 STOS pc 51 9001 022 9 l r Pe bli quy zuy yurod uon lul su l l yq p r woy JUU III asuejsip sdajs aw ajqeuea sey uoles SJUSWILUOD fEUODn PPy d PUD aquiny s bnee zods Arewuwins eyg Mold issaippy auoydajay 290 u02 Auedwoy Auedwoy Bunsejyjuoy Palod SILI 0 101221140 LEE 5500 Hf PANBPISIO SHIPS s1299 papoys a48 DROP mdu d ng Spl l TWNOILdO 404 5192 papoys pas UNM M N Splal4 AGOLYONV AL 404 06 MI z DPE S87 MI danat 591 118 622 danat Poz 282 89T WI UIEN Tov ul aso ae Pu PuHL9AlM ku ya auery meyar He ese anug ECEE ieee sap SOUIIBS WA SAUNA SINIL OND 33 WBA Boo eeu see retin pee Sasi Gore E9999 Fi Glib ieee seq S6067 fk yy Dtf BILAOBING St th ADN 4467 MAMA pues pp 66 60 81 yepdfi 3857 UOJSIIA 9724 LAOdHAA TVOINHOAL AVA 19140 di samen arog aq Smog par b z III AL AM NDUNI 2202 danao y danav y XAOSU CI NV eNOS ih Taquiny SJUIOg BULOWUOW JO 3517 suone S jo on suone s BULIOJIUON papug eeg paye s Bulb5o7 eeg BulBBo7 eeg poypay eeg mun sixy X JO
52. veral Sites Fundamental analysis of shear dispersion was undertaken by Taylor 1954 His work provides a theoretical framework for the analysis and prediction of longitudinal dispersion although in river studies compliance with the basic assumptions is rarely achieved More recently the Aggregated Dead Zone model Beer and Young 1984 has been used to quantify and describe the observed features from river tracer studies Whilst both these techniques may be applied to predict the effect of pollution spillages they do require the use of modelling tools and a knowledge of the parameters A brief description of both techniques 15 provided in the following sections On occasions it may be necessary to provide predictions for a river reach where neither a model nor a knowledge of the required parameters are available For such situations alternative approaches have been proposed These range from simply relating the solute velocity v to a power x of the river discharge Q of the form v oc Q to a technique described by Jobson 1997 working for the United States Geological Survey USGS R amp D TECHNICAL REPORT P346 3 Jobson 1997 described a predictive method based on information which should be available at the time of an incident t utilises an extensive data set to estimate the rate of movement of solutes through river reaches the rate of attenuation of the peak concentration of a conservative solute with time and the length of tim
Download Pdf Manuals
Related Search