Delft3D-PART User Manual

Water quality and aquatic ecology modelling suite
D-Water Quality
T
AF
DR
Particle tracking
User Manual
DR
AF
T
T
AF
D-WAQ PART
Simulation of mid-field water quality and oil spills, us-

DR
ing particle tracking
User Manual
D-Water Quality
Version: 2.15
SVN Revision: 77723
13 April 2023
D-WAQ PART, User Manual
T
AF
DR
Published and printed by:

Deltares telephone: +31 88 335 82 73
Boussinesqweg 1 e-mail: info@deltares.nl
2629 HV Delft www: https://www.deltares.nl
P.O. 177
2600 MH Delft
The Netherlands
For sales contact: For support contact:

telephone: +31 88 335 81 88 telephone: +31 88 335 81 00
e-mail: software@deltares.nl e-mail: software.support@deltares.nl
www: https://www.deltares.nl/software www: https://www.deltares.nl/software
Copyright © 2023 Deltares

All rights reserved. No part of this document may be reproduced in any form by print, photo
print, photo copy, microfilm or any other means, without written permission from the publisher:
Deltares.
Contents
Contents
List of Tables vii
List of Figures ix
1 A guide to this manual 1

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Manual version and revisions . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.3 Typographical conventions . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.4 Changes with respect to previous versions . . . . . . . . . . . . . . . . . . 2
T
2 Introduction to D-WAQ PART 5
2.1 Short description of the program functionality . . . . . . . . . . . . . . . . . 5
2.2 Specification of the program package . . . . . . . . . . . . . . . . . . . . . 7
2.3 Installing the program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
AF
2.4 Error checking . . . . . . . . . . . . . . .
2.5 Particle tracking module within Delft3D . . .
2.6 Input data files . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
8
8
9
2.7 Limitations in PART-GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3 Getting started 11
3.1 Starting Delft3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2 Particle tracking module . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.3 Change working directory . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.4 Coupling module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
DR
3.4.2 Coupling GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

3.4.3 Running the coupling module . . . . . . . . . . . . . . . . . . . . . 18
3.5 Starting PART-GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.5.1 Accessing the data groups . . . . . . . . . . . . . . . . . . . . . . 20
3.5.2 Visualisation Area window . . . . . . . . . . . . . . . . . . . . . . 20
3.5.3 Editing tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.5.4 Saving a D-WAQ PART input file . . . . . . . . . . . . . . . . . . . 21
3.6 Quitting the program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4 Graphical User Interface 23

4.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1.1 Initial conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1.2 Boundary conditions . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1.3 Co-ordinate system . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1.4 Data groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.2 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.3 Hydrodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.3.1 Number of layers in hydrodynamics . . . . . . . . . . . . . . . . . . 24
4.4 Substances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.4.1 Tracers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.4.2 Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.5 Time frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
4.5.1 Date-time format . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
4.5.2 Start time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.5.3 Stop time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.5.4 Time step . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.5.5 Reference time input . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.6 Initial condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Deltares iii
4.7 Numerical parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

4.7.1 Number of particles . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4.7.2 Vertical dispersion . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.8 Instantaneous releases . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.8.1 Co-ordinates of instantaneous release location . . . . . . . . . . . . 33
4.8.2 Particles used . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.8.3 Radius of instantaneous release . . . . . . . . . . . . . . . . . . . 34
4.8.4 Release time and mass of release . . . . . . . . . . . . . . . . . . 35
4.9 Continuous releases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.9.1 Co-ordinates of continuous release location . . . . . . . . . . . . . . 36
T
4.9.2 Particles used . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.9.3 Radius of continuous release . . . . . . . . . . . . . . . . . . . . . 37
4.9.4 Release time and rate . . . . . . . . . . . . . . . . . . . . . . . . 38
4.9.5 Concentration of continuous release . . . . . . . . . . . . . . . . . 38
4.10 Process parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
AF 4.10.1 Decay rates . . . . . . . . . . . . . . . . . . . . . . . .
4.10.2 Sedimentation/erosion . . . . . . . . . . . . . . . . . .
4.10.3 Physical parameters . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
40
41
43
4.10.4 Oil Model parameters . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.11 Observation points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.12 Output options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.12.1 Map file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.12.2 Zoom grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.12.3 Particle tracks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.12.4 Output file types . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.13 Executing the scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
DR
4.13.1 Executing the water quality mid-field module D-WAQ PART . . . . . . 54

4.13.2 Estimating run time . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.13.3 File sizes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5 Tutorial 55
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
5.2 D-WAQ PART tutorial cases . . . . . . . . . . . . . . . . . . . . . . . . . . 55
5.2.1 Tutorial cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
5.2.2 Starting D-WAQ PART . . . . . . . . . . . . . . . . . . . . . . . . 56
5.3 Coupling the hydrodynamic results . . . . . . . . . . . . . . . . . . . . . . 56
5.3.1 Coupling module . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.3.2 Definition of the input . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.3.3 Saving input and running the coupling module . . . . . . . . . . . . 60
5.4 Case 1: Tracer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.4.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.4.2 Hydrodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.4.3 Substances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.4.4 Time frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.4.5 Numerical parameters . . . . . . . . . . . . . . . . . . . . . . . . 65
5.4.6 Instantaneous releases . . . . . . . . . . . . . . . . . . . . . . . . 66
5.4.7 Continuous releases . . . . . . . . . . . . . . . . . . . . . . . . . 69
5.4.8 Process parameters . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.4.9 Observation points . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.4.10 Output options . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.4.11 Saving the input . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.5 Case 2: Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.5.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.5.2 Hydrodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
iv Deltares
Contents
5.5.3 Substances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.5.4 Time frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.5.5 Initial condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.5.6 Numerical parameters . . . . . . . . . . . . . . . . . . . . . . . . 84
5.5.7 Instantaneous releases . . . . . . . . . . . . . . . . . . . . . . . . 85
5.5.8 Continuous releases . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.5.9 Process parameters . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.5.10 Observation points . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.5.11 Output options . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5.5.12 Saving the input . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
T
5.6 Executing a scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
5.7 Viewing the report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
5.8 Viewing the results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
6 Conceptual model 101

AF
6.1 Introduction . . . . . . . . . . . . . . . . . . . .
6.1.1 Operational functionality . . . . . . . . .
6.1.2 Physical system . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 101
. 101
. 101
6.1.3 Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.2 Principles of a random walk particle tracking model . . . . . . . . . . . . . . 102
6.3 Dispersion coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
6.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
6.3.2 Horizontal dispersion . . . . . . . . . . . . . . . . . . . . . . . . . 102
6.3.3 Vertical dispersion . . . . . . . . . . . . . . . . . . . . . . . . . . 104
6.4 3 Dimensional flow field . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.5 Oil spill module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
DR
6.6 Data requirements for the conceptual model . . . . . . . . . . . . . . . . . 114
7 Algorithmic implementation 117

7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.2 Advection schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.3 Dispersion schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
References 119
A Input files D-WAQ PART 121

A.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
A.2 General set-up of the input file . . . . . . . . . . . . . . . . . . . . . . . . 121
A.3 Oil module constants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
A.4 Initial condition file for the oil module . . . . . . . . . . . . . . . . . . . . . 130
A.5 Polygon file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
A.6 Initial condition file for the oil module . . . . . . . . . . . . . . . . . . . . . 133
Deltares v
T
AF
DR
vi Deltares
List of Tables
List of Tables
2.1 Limitations of the PART-GUI . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.1 Specifications of the tutorial cases . . . . . . . . . . . . . . . . . . . . . . 55

5.2 Timers for the Tracer tutorial case . . . . . . . . . . . . . . . . . . . . . . . 64
5.3 Details of dye release 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.5 Details of the continuous release . . . . . . . . . . . . . . . . . . . . . . . 69
5.7 Process parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.8 Time breakpoints definition . . . . . . . . . . . . . . . . . . . . . . . . . . 77
T
5.9 Timers for the Oil tutorial case . . . . . . . . . . . . . . . . . . . . . . . . 82
5.10 Details of oil releases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.12 Process parameters for Sedimentation and Erosion . . . . . . . . . . . . . . 89
5.13 Physical process parameters . . . . . . . . . . . . . . . . . . . . . . . . . 90
5.14 Process parameters for oil . . . . . . . . . . . . . . . . . . . . . . . . . . 91
AF
5.15 Time breakpoints definition . . . . . . . . . . . . . . . . . . . . . . . . . . 94
DR
Deltares vii
T
AF
DR
viii Deltares
List of Figures
List of Figures
2.1 Example of a D-WAQ PART oil spill simulation . . . . . . . . . . . . . . . . 6
2.2 Software architecture around the particle tracking module . . . . . . . . . . . 8
3.1 Main window of Delft3D-MENU . . . . . . . . . . . . . . . . . . . . . . . . 11

3.2 Particle tracking menu in Delft3D-MENU . . . . . . . . . . . . . . . . . . 12
3.3 Select working directory window . . . . . . . . . . . . . . . . . . . . . . 12
3.4 Repeating a tidal cycle of 12 hours stored on the communication file, for a
D-WAQ PART simulation of 36 hours . . . . . . . . . . . . . . . . . . . . . 14
T
3.5 Principle of vertical aggregation where 10 layers in the hydrodynamic simula-
tion are aggregated to 5 layers in the water quality simulation . . . . . . . . . 15
3.6 Hydrodynamic coupling selection window . . . . . . . . . . . . . . . . . . 15
3.7 COUP-GUI window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.8 Hydrodynamics sub-window in COUP-GUI . . . . . . . . . . . . . . . . . . 17
AF
3.9
3.10
3.11
Vertical aggregation using the Layer editor . . . . . . . . . . . . . . . . .
Main menu of PART-GUI . . . . . . . . . . . . . . . . . . . . . . . . . .
Hydrodynamics Data Group . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
18
19
20
3.12 Selecting the Visualisation Area window . . . . . . . . . . . . . . . . . . . 20
3.13 Visualisation Area window . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.14 Table editing options in the PART-GUI main menu bar . . . . . . . . . . . . . 21
3.15 Saving the input file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.16 Exiting the program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.17 Window to save <∗.mdp> file before quitting . . . . . . . . . . . . . . . . . 22
4.1 Data Group Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

DR
4.2 Data Group Hydrodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . 24

4.3 Navigation window to select a hydrodynamic scenario . . . . . . . . . . . . . 25
4.4 Data Group Substances when Tracers are selected . . . . . . . . . . . . . . 25
4.5 Confirmation window to delete substance . . . . . . . . . . . . . . . . . . . 27
4.6 Data Group Substances when Type of model Oil is selected . . . . . . . . . 27
4.7 Time Frame Data Group . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
4.8 Data Group Initial Condition . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.9 Data Group Numerical Parameters . . . . . . . . . . . . . . . . . . . . . . 30
4.10 Instantaneous Releases Data Group for tracer substances . . . . . . . . . . 32
4.11 Release tables for instantaneous releases window for defining release time
and mass of release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.12 Instantaneous Releases Data Group for oil substances . . . . . . . . . . . . 34
4.13 Continuous Releases Data Group for tracer substances . . . . . . . . . . . . 36
4.14 Release tables for continuous releases window for defining release time,
discharge and concentration of release . . . . . . . . . . . . . . . . . . . . 37
4.15 Process parameters Data Group . . . . . . . . . . . . . . . . . . . . . . . 39
4.16 Process Parameters Data Group for oil modelling . . . . . . . . . . . . . . . 39
4.17 Decay rates sub-window . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.18 Sedimentation/erosion sub-window . . . . . . . . . . . . . . . . . . . . . 41
4.19 Sediment/Erosion constants sub-window . . . . . . . . . . . . . . . . . . 41
4.20 Settling velocities sub-window . . . . . . . . . . . . . . . . . . . . . . . . 42
4.21 Physical parameters sub-window . . . . . . . . . . . . . . . . . . . . . . 44
4.22 Oil Model input for the Oil model . . . . . . . . . . . . . . . . . . . . . . . 46
4.23 Observation Points Data Group . . . . . . . . . . . . . . . . . . . . . . . . 49
4.24 Output Options Data Group . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.25 Definition of Zoom grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.26 Defining zoom grid using the Visualisation Area window . . . . . . . . . . . 52
4.27 Definition of Zoom grid output time intervals . . . . . . . . . . . . . . . . . . 52
Deltares ix
5.1 Coupling GUI - Data Group Description . . . . . . . . . . . . . . . . . . . 57

5.2 Vertical aggregation using the Layer editor. Left not aggregated and right
aggregated layers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
5.3 Coupling GUI - Data Group Hydrodynamics . . . . . . . . . . . . . . . . . 59
5.4 Coupling GUI - Data Group Dispersion . . . . . . . . . . . . . . . . . . . . 60
5.5 Coupling report output window . . . . . . . . . . . . . . . . . . . . . . . . 60
5.6 Selection window for Particle tracking . . . . . . . . . . . . . . . . . . . . 61
5.7 Main window of the PART-GUI . . . . . . . . . . . . . . . . . . . . . . . . 62
5.9 Data Group Hydrodynamics showing the selected hydrodynamic results <com-
T
tut_fti_part.hyd> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.10 Data group Substances . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.11 Data Group Time Frame. . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
5.12 Data Group Numerical Parameters after editting . . . . . . . . . . . . . . . 65
5.13 Data Group Instantaneous Releases . . . . . . . . . . . . . . . . . . . . . 67
AF5.14
5.15
Release tables for instantaneous releases window of a selected instanta-
neous release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Visualisation Area window . . . . . . . . . . . . . . . . . . . . . . . . . .
68
68
5.16 Data Group Continuous Releases . . . . . . . . . . . . . . . . . . . . . . . 70
5.17 Release tables for continuous releases window of continuous release 1 . . 70
5.18 Data Group Process Parameters with sub-data group Decay rates . . . . . . 71
5.19 Sub-data group Sedimentation/erosion – Constants . . . . . . . . . . . . . . 72
5.20 Settling Velocities details (A0 list box selected) . . . . . . . . . . . . . . . . 73
5.21 Physical Process parameters . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.22 Data Group Observation Points . . . . . . . . . . . . . . . . . . . . . . . . 75
5.23 Data Group Output Options, with also showing options for Grid Definition of
DR
the Zoom grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

5.24 Zoom Grid Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.25 Selection window for Particle tracking . . . . . . . . . . . . . . . . . . . . 78
5.26 Main window of the PART-GUI . . . . . . . . . . . . . . . . . . . . . . . . 79
5.28 Data Group Hydrodynamics showing the selected hydrodynamic results <com-
tut_fti_part.hyd> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.29 Data Group Substances . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.30 Data Group Time Frame . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.31 Data Group Initial Condition . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.32 Selection window for the initial conditions file . . . . . . . . . . . . . . . . . 83
5.33 Data Group Numerical Parameters . . . . . . . . . . . . . . . . . . . . . . 84
5.34 Data Group Instantaneous Releases . . . . . . . . . . . . . . . . . . . . . 85
5.35 Release tables for instantaneous releases window for oil patch 1 . . . . . . 86
5.36 Release tables for instantaneous releases window for oil patch 2 . . . . . . 86
5.37 Data Group Process parameters with sub-data group Decay rates . . . . . . 88
5.38 Sub-data group Sedimentation/Erosion . . . . . . . . . . . . . . . . . . . . 88
5.39 Settling Velocities details (A0 selected in list box) . . . . . . . . . . . . . . . 89
5.40 Physical Process parameters . . . . . . . . . . . . . . . . . . . . . . . . . 90
5.41 Oil Model specific parameters . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.42 Data Group Observation Points . . . . . . . . . . . . . . . . . . . . . . . . 92
5.43 Data Group Output Options, also showing options for Definition of Zoom grid . 93
5.44 Zoom Grid Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
5.45 Tracer Case: Instantaneous release: situation at 18:00 hrs (upper) and 19:30
hrs (lower) in the surface layer . . . . . . . . . . . . . . . . . . . . . . . . 96
5.46 Tracer Case: Continuous release: situation at 20:30 hrs in layer 3: MAP-file
(upper) and PLO-file (lower) . . . . . . . . . . . . . . . . . . . . . . . . . . 97
x Deltares
List of Figures
5.47 Oil Case: Situation at 19:30 hrs in the surface layer: Ekofisk dispersed (upper)
and Floating HFO (lower) . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
5.48 Oil Case: Situation at 17:00 hrs (upper) and 19:30 hrs (lower), PLO-file for
Floating HFO in the surface layer . . . . . . . . . . . . . . . . . . . . . . . 99
6.1 Vertical disperison coefficient and its gradient as a function of relative depth
for the k -L turbulence model . . . . . . . . . . . . . . . . . . . . . . . . . 105
6.2 Relationship between viscosity and the dispersion parameter C0 (Delvigne
and Hulsen, 1994) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
T
AF
DR
Deltares xi
T
AF
DR
xii Deltares
1 A guide to this manual
1.1 Introduction
In order to make this manual more accessible, a brief description of the contents of each of
the chapters will be given here. If you are a novice user of D-WAQ PART, you are advised to
work through the tutorial that is provided in Chapter 5. This should be done in combination
with Chapter 4, where the graphical user interface options are explained in detail.
This document describes the operation of PART within the Delft3D suite.
T
⋄ Chapter 2: Introduction to D-WAQ PART, a short overview of the program is given
together with some specifications of the program package.
⋄ Chapter 3: Getting started, the general functionality of the D-WAQ PART framework is
described, together with a brief description of how to start and quit the program.
⋄ Chapter 4: Graphical User Interface, provides a detailed description of the Graphical
AF User Interface (PART-GUI).
⋄ Chapter 5: Tutorial, provides you with step-by-step instructions to complete a “tracer”
and an “oil” tutorial case. Both result in a complete input file and can be executed. Output
from the tutorial cases can be compared with the results provided. The emphasis here is
to provide practical experience in the use of the module.
⋄ Chapter 6: Conceptual model, gives an overview of the modelling concepts used in
D-WAQ PART.
⋄ Chapter 7: Algorithmic implementation, describes the algorithms for the advection and
dispersion schemes.
⋄ References, provides the references used in this manual.
DR
⋄ Appendix A: Input files D-WAQ PART, a description of the ASCII input file (generated
by PART-GUI) is given. The ASCII input file can be edited by advanced users who want to
apply functionalities that are not (yet) supported by the GUI.
1.2 Manual version and revisions

A manual applies to a certain release of the related numerical program. This manual, version
2.14, applies to D-WAQ PART version 3.73.00, and PART-GUI version 2.26.30.
1.3 Typographical conventions

Throughout this manual, the following conventions help you to distinguish between different
styles of text used to describe various aspects of the graphical user interface.
Example Description
Module Title of a window or a sub-window are in given in bold.

Project Sub-windows are displayed in the Module window and
cannot be moved.
Windows can be moved independently from the Mod-
ule window, such as the Visualisation Area window.
Deltares 1 of 137
Example Description
Save Item from a menu, title of a push button or the name of

a user interface input field.
Upon selecting this item (click or in some cases double
click with the left mouse button on it) a related action
will be executed; in most cases it will result in displaying
some other (sub-)window.
In case of an input field you are supposed to enter input
T
data of the required format and in the required domain.
<\tutorial\wave\swan-curvi> Directory names, filenames, and path names are ex-

<siu.mdw> pressed between angle brackets, <>. For Linux en-
vironments a forward slash (/) is used instead of the
backward slash (\) for Windows environments.
AF “27 08 1999” Data to be typed by you into the input fields are dis-
played between double quotes.
Selections of menu items, option boxes etc. are de-
scribed as such: for instance ‘select Save and go to
the next window’.
delft3d-menu Commands to be typed by you are given in the font

Courier New, 10 points.
In this User manual, user actions are indicated with this

DR
arrow.
[m s−1 ] [−] Units are given between square brackets when used
next to the formulae. Leaving them out might result
in misinterpretation. Most units will be in SI notation.
[m AD] stands for ’meter Above Datum’, which denotes
a level relative to the vertical reference system in the
model.
1.4 Changes with respect to previous versions
Version Description
2.15 Update of Friesian Tidal Inlet tutorial model<∗.mdp> file
2.14 Introduction of the <∗.mdp> file
2.13 PART v3.72.00 and higher can handle thin dams from FLOW. Thus:
In section 4.3 Remark removed to warn that thin dams defined in the FLOW
model are not recognised in the PART model.
In Chapter 5: data files <com-s33.∗> replaced by <com-f35_part.∗>,
file <s33_oil.inp> replaced by <s35_oil.inp>; and <s33_tracer.inp> by
<s35_tracer.inp>. Finally, figures updates.
continued on next page
2 of 137 Deltares
A guide to this manual
– continued from previous page
Version Description
2.12 MENU screens updated.

Defining coupling input file and starting the coupling now separate MENU op-
tions.
Remark added about incorrect interpolation of wind direction if the wind passes
the North (0 degrees).
T
Remark added to warn that thin dams defined in the FLOW model are not recog-
nised in the PART model.
2.11 The most urgent change that was implemented in the model was the correction
of the entrainment of oil in water. Because emulsification was considered es-
sential for the simulation of the fate of oil, this process was also included. Since
AF interaction of processes are significant, a number of feedback mechanisms were
also implemented. After implementation, the changes in the specification of the
modified compared with the standard oil module is as follows:
⋄ Viscosity is now an input parameter and hence an oil characteristic, whilst
the entrainment parameter (C0) is now calculated using a relationship that
was derived from published data by Delvigne and Hulsen (1994).
⋄ Viscosity is not a fixed parameter but a variable, depending on the evapora-
tion and emulsification.
⋄ Dispersion (entrainment) process is now consistent with the ADIOS model
and independent of the grid.
DR
⋄ Emulsification is implemented in the model.

⋄ The emulsification is initiated when the evaporated fraction exceeds a given
fraction of the oil (input parameter by the user and oil type dependent).
⋄ Emulsification is calculated by the increase in the water content of the oil
and affects oil viscosity.
⋄ Since emulsification affects oil viscosity, entrainment parameter (C0) is af-
fected by a change in viscosity and the entrainment flux reduces when vis-
cosity increases.
⋄ Evaporation now affects oil viscosity and hence oil dispersion (entrainment).
Evaporation is implemented using the analytical solution of a first order pro-
cess.
⋄ A volatile fraction can be specified. The non-volatile fraction will not be sub-
ject to evaporation.
2.11 Emulsification affects evaporation by using the water content of the oil as calcu-
lated during emulsification. When the water content reaches its defined maxi-
mum, evaporation ceases.
All variables take account of the age of the particles and are being tracked for
each particle individually, thus the processes are applied correctly for continuous
discharges or several instantaneous discharges at different times.
2.11 section 4.12.2: The use of a Zoom grid is compulsory.
2.00 Improved layout of manual.
Deltares 3 of 137
T
AF
DR
4 of 137 Deltares
2 Introduction to D-WAQ PART
2.1 Short description of the program functionality

The PART module of Delft3D simulates transport and simple water quality processes by
means of a particle tracking method using the (2 or 3-dimensional) flow data from the FLOW
module. The tracks are followed in three dimensions over time, whereby a dynamic concen-
tration distribution is obtained by calculating the mass of particles in the model grid cells.
The processes are assumed to be deterministic except for a random displacement of the
T
particle at each time step. The particle tracking method is based on a random-walk method
since the simulated behaviour is stochastic and the number of particles is limited (Rubinstein,
1981). (It is worth noting that D-WAQ PART is the only stochastic model in the whole Delft3D
suite.)
AF Particle tracking allows water quality processes to be described in a detailed spatial pattern,
resolving sub-grid concentration distributions. D-WAQ PART is best suited for studies over
the mid-field range (200 m – 15 km) of instantaneous or continuous releases, simulation of
a plume (for example oil-spill), and modelling of the transport of substances such as salt,
bacteria, rhodamine dye, oil, BOD, or other conservative or decaying chemical substances
(following first-order kinetics). For this reason, the D-WAQ PART module is often referred to
as the Mid-field Water Quality module.
In D-WAQ PART, two modules are available:
⋄ Tracer module: simulation of conservative or first order decaying substances.

DR
⋄ Oil spill module: simulation of oil spills with floating and dispersed oil fractions.
Physical components
The physical components of the D-WAQ PART model system include:
⋄ the water system: a lake, estuary, harbour or river, possibly with open boundaries to other
water systems. Tidal variations are included.
⋄ discharges due to human activities that may be instantaneous and/or continuous;
⋄ chemical substances like rhodamine dyes, salt, oil or suspended solids;
⋄ time varying wind fields;
⋄ settling and erosion of suspended matter;
⋄ concentration dependent settling velocity.
Releases may be defined as instantaneous releases (e.g. dye releases) or continuous re-
leases (e.g. from sewage outfalls).
Deltares 5 of 137
T
Figure 2.1: Example of a D-WAQ PART oil spill simulation
AF
Physical processes
Physical processes or phenomena that the model was designed to represent are:
⋄ the dynamics of patches close to an outfall location;

⋄ simple first-order decay processes like the decay of several fractions of oil;
⋄ vertical dispersion for well-mixed systems;
⋄ horizontal dispersion due to turbulence. According to turbulence theory this dispersion
increases in time.
⋄ the effects of time-varying wind fields on the patches;
⋄ the effects of bottom-friction on the patches;
DR
⋄ the existence of a plume at the outfall (rather than a point-source) by starting the simulation
from a circular plume with an estimated or field-measured radius.
⋄ settling of particles, where a concentration dependent settling, subject to a minimum and
maximum settling velocity, can be specified;
⋄ transport and fate of spilled oil. Processes that are included in the oil module are: advec-
tion of floating oil by wind and currents, dispersion (entrainment in water) of oil induced
by wind waves (depending on wind speed and oil characteristics), evaporation of floating
oil, emulsification, decay, and sticking of oil to the coastline or seabed. Changes of oil
properties (density, viscosity, water content) due to these processes are included in the oil
module. Additional advection due to wind drag is implemented in 3D (for 2D the logarith-
mic profile of the current speed results in a similar wind driven transport), whilst in 3D a
deflection angle is included to represent the Coriolis effect of wind induced advection due
to waves.
⋄ the model may be started from a known initial distribution of oil, e.g. a remote sensing
image of an oil spill.
Applications
The model may be applied to various problems with the physical components, substances
and processes described above. However, the time-scale of simulations is usually limited to
a few weeks, in order to simulate accurately using a large number of particles. The model is
very often applied for plume (outfall) studies with spatial scales of a few kilometres. In such
cases, the hydrodynamic grid is often too coarse for other modelling techniques. The particle
tracking method is not limited by the spatial resolution of the hydrodynamic database. Thus,
detailed outfall studies can be done even within one computational grid cell covering a large
area. A costly refinement of the grid in this area is not required.
6 of 137 Deltares
Introduction to D-WAQ PART
How does particle modelling work?

The position of every individual particle can be influenced by:
⋄ advection (transport by water flow)
⋄ diffusion/dispersion (a random component)
⋄ settling (including sedimentation/erosion characteristics)
In D-WAQ PART, every particle is subject to a random displacement. The distance of the
displacement depends on the local dispersion and the time step in the model. The direction
T
of the displacement is random. The method used is often referred to as the ‘random walk
method’ or the ‘Monte Carlo method’.
The mass of a particle represents the amount of a substance attached to it. The mass of a
particle can be influenced by:
AF ⋄ decay rate (first order decay)
⋄ evaporation (of oil).
In D-WAQ PART, the position of every individual particle (x, y and z co-ordinate) is known
from its release until the end of the simulation. The position of a particle does not depend on
the grid (size) used and so sub-grid concentration patterns can be calculated.
A release of a substance is distributed over the number of particles specified in the model
run. For example, if a dye release of 10 000 particles is used to represent 100 grams of
tracer, then each particle represents 0.01 g of tracer. In a continuous release the specified
DR
number of particles is distributed over the complete simulation time. That is, if a total number
of 10 000 particles are released in 1 day, this is equivalent to 0.116 particles per second.
Furthermore, if each particle is assigned 10 g of tracer, this is equivalent to a continuous
release of 1.16 g/s.
The concentration of a substance is determined by the number of particles per unit volume.
The volume used to calculate the concentrations is defined in one of two ways:
⋄ The hydrodynamic segments (from the Delft3D-FLOW computation).
⋄ Segments of a counting grid, the so-called zoom grid, which can be located in any position
in the computational domain. The zoom grid defines a regular, rectangular grid in which
the rectangles are usually much smaller than the hydrodynamic segments.
The model simply counts the number of particles present in each volume and calculates the
corresponding concentration of each substance.
2.2 Specification of the program package

This User Manual describes the use of the Graphical User Interface (PART-GUI) for the com-
putational model D-WAQ PART of the Delft3D suite. The PART-GUI enables you to input and
to retrieve data necessary for preparation and execution of a modelling run.
The manual also describes input and output files generated by PART-GUI and D-WAQ PART.
Deltares 7 of 137
2.3 Installing the program

The installation of the Delft3D suite is described in the Installation Manual that is supplied with
the software.
2.4 Error checking

The syntax of the input file is checked by the computational core of D-WAQ PART. Syntax
errors cannot occur if you use the PART-GUI. However, if you edit the input files directly,
syntax errors may occur and you are responsible for tracing the origin of the error.
T
The program may stop without an error message on the screen. Error messages are always
given in the report file <∗.out> of the simulation. Looking for the word ‘ERROR’ (capital
AF letters) in this report file can be very useful.
2.5 Particle tracking module within Delft3D

PART-GUI enables you to interact with the mid-field water quality model D-WAQ PART. The
other GUI modules like FLOW-GUI, WAQ-GUI, etc. enable you to interact with other modules
in the Delft3D modelling system. These interactions include the input and retrieval of data that
is necessary for preparation, execution and post-processing of a modelling run.
Delft3D-MENU
Hydrodynamics Water Quality Particle Tracking ...

DR
coupling PART-GUI Delft3D-PART QUICKPLOT

module (visualisation)
Figure 2.2: Software architecture around the particle tracking module
Figure 2.2 shows a part of the Delft3D suite software structure and the location of the Parti-
cle Tracking module with respect to the other Delft3D modules (e.g. Hydrodynamics, Water
Quality, etc.). The most important components in this figure are:
⋄ coupling module
⋄ graphical user interface (PART-GUI)
⋄ particle tracking computational module (D-WAQ PART)
⋄ GPP and QUICKPLOT modules for visualisation of results
Descriptions of the other Delft3D modules can be found in the associated User Manuals
Deltares (2013, 2023); WAVE UM (2013); GPP UM (2013).
8 of 137 Deltares
Introduction to D-WAQ PART
2.6 Input data files

Execution of the D-WAQ PART computational module depends on the presence of a number
of D-WAQ PART input files. PART-GUI enables you to input data that are necessary for writing
an input file and to retrieve data from existing input files. A D-WAQ PART scenario is defined
by the <∗.mdp> file (mdp: Master Definition Part), the corresponding <∗.inp> and the
<∗.hyd> file. For each new scenario the files <∗.mdp> and <∗.inp> are created by PART-
GUI. The files <∗.mdp> and <∗.inp> are required when opening an existing scenario. The
<∗.hyd> file is created by the Coupling module.
T
The input data are divided in data groups: Description, Hydrodynamics, Substances, Time
Frame, Initial Condition, Numerical Parameters, Instantaneous Releases, Continuous Re-
leases, Process Parameters, Observation Points and Output Options.
Coupling module
AF The coupling module provides D-WAQ PART with all of the necessary files from the hy-
drodynamic (Delft3D-FLOW) modelling results. The hydrodynamic results are described in
a <∗.hyd> file and the corresponding data is given in the associated grid files (<∗.lga>,
<∗.lgt>, <∗.cco>), volumes file (<∗.vol>) and flows file (<∗.flo>).
2.7 Limitations in PART-GUI

In this section the list of (known) limitations of PART-GUI is described. Most of them have
to do with a maximum number of substances, observations points etc. Please take them into
account while using the PART-GUI.
DR
Table 2.1: Limitations of the PART-GUI
Number Description Value
1 Maximum number of substances (oil counts as 3 substances) 90
2 Maximum number of continuous releases (note: number of instanta- 25

neous releases is unlimited)
3 Maximum number of columns in (any) GUI table 70
4 Maximum number of oil types (oil counts as “1”) 30
5 Maximum number of time breakpoints in GUI table 1 000
6 Maximum number of layers in case of PLO output 100
Deltares 9 of 137
T
AF
DR
10 of 137 Deltares
3 Getting started
3.1 Starting Delft3D

The Delft3D suite is a range of modules that can be run independently of one another. The
main menu of Delft3D provides access to each of the main modules. Delft3D-MENU can be
started either:
⋄ In Microsoft®Windows, select Delft3D in the Programs Menu or click on the Delft3D icon
on the desktop.
⋄ On Linux/UNIX, type delft3d-menu on the command line.
T
The main window of Delft3D-MENU is shown in Figure 3.1.
3.2
AF Particle tracking module
The PART module is started by selecting Start in the main menu. The Particle tracking
selection menu is shown in Figure 3.2.
From the Particle tracking menu, it is possible to start the Coupling module, start the graphi-
cal user interface (PART-GUI), begin computations with D-WAQ PART, view output report files
(ASCII) and to visualise the results using GPP or Delft3D-QUICKPLOT.
3.3 Change working directory

It is also possible to select and/or change the default working directory in which the input are
DR
located and to which the output files will be written.
The Select working directory button opens a window in which the current working directory is
identified as shown in Figure 3.3.
The current working directory can be changed by browsing to and entering the preferred
directory using the window of Figure 3.3 and confirming with OK. You have to ensure that you
have write permission in the selected directory.
Figure 3.1: Main window of Delft3D-MENU
Deltares 11 of 137
T
AF
Figure 3.2: Particle tracking menu in Delft3D-MENU
DR
Figure 3.3: Select working directory window
12 of 137 Deltares
Getting started
3.4 Coupling module
3.4.1 Introduction
The transport of substances is based on the hydrodynamic results from a Delft3D-FLOW com-
putation. Delft3D-FLOW generates a so-called communication file, consisting of a data file
and a definition file. The names of these files are <com-∗.dat> and <com-∗.def> respec-
tively, where ∗ indicates a run identification of the Delft3D-FLOW simulation. These files need
to be converted to a so-called hyd-file, that can be used by D-WAQ PART. But it is preferred
now use the option to let Delft3D-FLOW write the hyd file directly (and skip the use of the
T
coupling program) by switching on ’Export WAQ input’ in the output section of Delft3D-FLOW.
The communication or hyd file should encompass a representative period of the hydrodynamic
results with a sufficient time-step resolution. Examples of a representative period would be a
single tide (e.g. ∼12.5 hours) or a spring-neap cycle (e.g. ∼15 days). The time-step resolution
AF must be selected to sufficiently describe the variations in the hydrodynamics. For example,
the time step for a repeating 12 hour tidal cycle should not be greater than one hour.
D-WAQ PART allows water quality processes to be described in a detailed spatial pattern
and can resolve sub-grid concentration distributions. The velocity field at the location of each
particle is interpolated between the velocity values given at each grid cell in the hydrodynamic
result file.
The data required to execute a D-WAQ PART simulation can be extracted from the hydrody-
namic communication file by the coupling module. In the coupling module, you must define
the:
DR
⋄ start time for the coupled hydrodynamics

⋄ stop time for the coupled hydrodynamics
⋄ time step for the coupled hydrodynamics
⋄ vertical aggregation of hydrodynamic results (if required)
But it is preferred now use the option to let Delft3D-FLOW write the hyd file directly by switch-
ing on ’Export WAQ input’ in the output section of Delft3D-FLOW. Use the second Output
panel to the start time, stop time, time step and optionally the vertical aggregation.
Remarks:
⋄ The coupling module is common to both D-WAQ PART and D-Water Quality. For this
reason, some of the options available in the coupling module are not applicable to a
D-WAQ PART simulation.
⋄ When you need a coupling from a domain decomposition hydrodynamic simulation you
have to define the input for each domain separately when using the com files (not
needed when using Export WAQ input).
⋄ When you run Delft3D-FLOW with MPI-based parallel, you cannot use the communi-
cation files. You will have to switch on the Export WAQ input in the output settings.
Use ’DD Start’ in the Water Quality or PART coupling menu with the MPI-based parallel
option to merge the output into one domain for D-Water Quality or D-WAQ PART.
⋄ Horizontal aggregation is not allowed when using D-WAQ PART, only vertical aggrega-
tion.
⋄ D-WAQ PART cannot work with Z -layer models because it assumes a uniform and
constant over time relative thickness of the vertical layers as stated in the hyd-file. For
Z -layer models, this not true. It also doesn’t check whether segments are below the
bed, so particles can end up below the bed.
Deltares 13 of 137
tidal cycle on communication file first repeated tidal cycle second repeated tidal cycle
0:00 6:00 12:00 18:00 0:00 6:00 12:00

1.2
0.8
0.4
w at e r l e ve l
0
-0.4
T
-0.8
-1.2
Figure 3.4: Repeating a tidal cycle of 12 hours stored on the communication file, for a
AF D-WAQ PART simulation of 36 hours
Rewinding hydrodynamic results

D-WAQ PART allows you to extend the particle tracking simulation beyond the period after
coupling the communication file. When D-WAQ PART reaches the end of the period in the
coupled hydrodynamics files, it will rewind the (coupled) hydrodynamic files and repeat them
(see Figure 3.4) until it reaches the end of the D-WAQ PART simulation.
Remark:
⋄ As D-WAQ PART will rewind the (coupled) hydrodynamic result, it is important that the
DR
start and the end of the representative period are similar. In Figure 3.4, the water level
is the same at the start and the end of this period. D-WAQ PART does not check this
closure error. You should check this yourself by plotting the change in water level due
to a rewind.
Aggregation
Hydrodynamic modelling usually requires a more detailed schematisation or grid than water
quality modelling. Therefore, it is often desirable to reduce the number of computational cells
in a water quality simulation to reduce computation time and the size of the output files. This
process is called aggregation. In D-WAQ PART, aggregation is only allowed in the vertical
direction (see Figure 3.5).
Remark:
⋄ Aggregation of the hydrodynamic results will inevitably result in a loss of information
and should be avoided if possible in D-WAQ PART simulations.
14 of 137 Deltares
Getting started
T
AF
Figure 3.5: Principle of vertical aggregation where 10 layers in the hydrodynamic simula-
tion are aggregated to 5 layers in the water quality simulation
DR
Figure 3.6: Hydrodynamic coupling selection window
Deltares 15 of 137
T
AF
Figure 3.7: COUP-GUI window
3.4.2 Coupling GUI

DR
If you have let Delft3D-FLOW write the hyd-file directly by switching on ’Export WAQ input’,
you can skip the rest of this section and go to section 3.5. Otherwise you will have to couple a
hydrodynamic result first. Select Coupling from the Particle tracking window (see Figure 3.2).
The Hydrodynamic coupling selection window is displayed, in which you can create or edit
the coupling input and start the coupling, see Figure 3.6.
Select Define input. The COUP-GUI window is opened as shown in Figure 3.7.
Only three data groups are selectable: Description, Hydrodynamics and Dispersion.
Description
Clicking Description opens a screen in which you can enter three lines of text. The lines can
be used to describe the coupling/aggregation. The maximum length of each of the lines is 40
characters.
16 of 137 Deltares
Getting started
T
AF
Figure 3.8: Hydrodynamics sub-window in COUP-GUI
Hydrodynamics
DR
Selecting Hydrodynamics opens the sub-window as shown in Figure 3.8.
First, you must select a hydrodynamics communication file by clicking Com-file under Hydro-
dynamics and grid frame. A file browser window is opened and you can navigate and select
the appropriate <com-∗.dat> file (the associated <com-∗.def> file is required to exist in the
same directory). After selecting the appropriate communication file, the times (start time, stop
time and time step) stored in the communication file are displayed. Next you must select ap-
propriate start and stop times and a time step for the coupled hydrodynamic results. The start
and stop times and the time step of the hydrodynamic results stored in the communication file
are shown next to the appropriate edit field.
The Start time, Stop time and Time step of the coupled hydrodynamics have to satisfy the
following conditions:
⋄ The start time of the coupled hydrodynamics has to be equal to or later than the start time
in the communication file.
⋄ The start time of the coupled hydrodynamics has to be before the stop time in the com-
munication file.
⋄ The stop time of the coupled hydrodynamics has to be before or equal to the stop time in
the communication file.
⋄ The stop time of the coupled hydrodynamics has to be later than the start time in the
communication file.
⋄ The stop time in the coupled hydrodynamics has to be later than the start time in the
coupled hydrodynamics.
⋄ The time step in the coupled hydrodynamics file has to be equal to an integer multiple of
the time step in the communication file. For example, if the time step in the communication
file is 30 minutes, allowable time steps in the coupled hydrodynamics are 30 minutes, 1
hour, 90 minutes, 2 hours, etc.
Deltares 17 of 137
⋄ The total time between the start time and the stop time in the coupled hydrodynamics has
to be an integer multiple of the time step in the coupled hydrodynamics.
To vertically aggregate the layers, click Edit layers. The Layer editor window will open, list-
ing the number of layers in the hydrodynamic simulation. By default, each layer in the water
quality model (using the coupled hydrodynamics) is set equal to one layer in the hydrody-
namic simulation. Figure 3.9 shows the Layer editor for a hydrodynamic simulation with 5
layers. By entering an integer number in one of the input boxes, you indicate the number of
hydrodynamic layers that have to be included in the specific water quality layer. In Figure 3.9
T
(right hand side), the first layer in the water quality simulation contains the first and second
hydrodynamic layers. The second layer in the water quality simulation contains the third and
fourth hydrodynamic layers. The third layer in the water quality simulation contains the fifth
hydrodynamic layer. Entering a number greater than 1, means that water quality layers will
be removed from the bottom (note that layer 1 is the surface layer). Lowering the number
AF in a text box, means that water quality layers will be added at the bottom. The sum of the
hydrodynamic layers will always be equal to the original number of layers.
DR
Figure 3.9: Vertical aggregation using the Layer editor
The left screen shows the original 10 hydrodynamic layers. The right screen shows a vertical
aggregation into 5 layers that will be applied in the D-WAQ PART simulation.
Remark:
⋄ Zero cannot be entered in any edit field. If the number entered is higher than the number
of hydrodynamic layers available, a warning will be given and the number will be trun-
cated to the maximum allowed. A 2D hydrodynamic simulation cannot be aggregated
vertically.
Dispersion (not applicable for D-WAQ PART)

Clicking Dispersion opens a sub-window that has options that are only applicable for a D-
Water Quality simulation. This window should not be used for a D-WAQ PART simulation.
3.4.3 Running the coupling module

After defining the desired coupling of the hydrodynamic file <com-∗.dat>, you must save the
coupled hydrodynamics file <∗.hyd> by selecting File → Save or File → Save As. . . from the
18 of 137 Deltares
Getting started
T
AF
Figure 3.10: Main menu of PART-GUI
menu bar. The default name of the coupled hydrodynamics file is <com-∗.hyd>. Leave the
COUP-GUI, File → Exit; you will be back in the Hydrodynamic coupling selection window,
DR
Figure 3.6.
Remark:
⋄ The <∗.hyd> file and the coupled hydrodynamics are always saved in the working
directory.
Select Start to perform the hydrodynamic coupling. Information about the coupling will be
displayed in a window. Close this window by selecting File → Exit.
The same information on the coupling is written to the <couplnef.out> file in the working
directory. You should examine this file: select Return to go back to the Particle tracking
selection window, Figure 3.2. Select Reports and next Coupling to view the report file from
the hydrodynamic coupling.
3.5 Starting PART-GUI

PART-GUI is started by selecting the menu item Define input from the Particle tracking menu.
The main menu of PART-GUI is shown in Figure 3.10.
The menu bar at the top of the screen provides options for opening and saving files, for editing
tables, for visualisation of the grid layout and for obtaining help.
Deltares 19 of 137
T
AF
Figure 3.11: Hydrodynamics Data Group
3.5.1 Accessing the data groups

A data group is accessed by clicking on the corresponding button on the left-hand side of the
DR
PART-GUI window. For example, clicking Hydrodynamics opens the data group window as
shown in Figure 3.11.
Pressing other buttons in the main window will change the window to the appropriate settings
for each of the data groups. Changes made in any of the data group windows will be retained
by PART-GUI. You will be prompted to save this data in an mdp-file <∗.mdp>, with a reference
to the input file <∗.inp>, when the PART-GUI session is ended.
3.5.2 Visualisation Area window

In PART-GUI, the Visualisation Area window is used to visualise the layout of the hydrody-
namic grid and to assist you in defining the Zoom Grid Output in the Output Options. Also,
observation points, instantaneous and continuous releases can be added using this tool. The
Visualisation Area window can be started by selecting View → Visualisation Area in the
menu bar as shown in Figure 3.12.
Figure 3.12: Selecting the Visualisation Area window
The Visualisation Area window then appears as shown in Figure 3.13. The hydrodynamic
computational grid is shown as black lines and the outline of the Zoom Grid Output is shown
as a red rectangle. The use of the Visualisation Area window to define the Zoom Grid Output
is described in the Output Options Data Group.
20 of 137 Deltares
Getting started
T
AF Figure 3.13: Visualisation Area window
3.5.3 Editing tables

In some data groups in PART-GUI, it is necessary to input data in tabular form (e.g. time series
data). The Table option in the menu bar allows you to Insert row, Copy row, Delete row or
Copy value as shown in Figure 3.14.
DR
Figure 3.14: Table editing options in the PART-GUI main menu bar
3.5.4 Saving a D-WAQ PART input file

In order to preserve changes and additions to the data groups, it is necessary to save the data
defined in the data groups in an input data file, <∗.mdp>.
You can save the input by clicking menu item File → Save (or Save as. . . ) in the menu bar
as shown in Figure 3.15. The file <runid.mdp> will be saved by PART-GUI and in the same
directory the file <runid.inp> will be written.
Deltares 21 of 137
T
Figure 3.15: Saving the input file
3.6 Quitting the program

A PART-GUI session is ended by selecting File → Exit from the menu bar as shown in Fig-
AF ure 3.16. The program will first ask whether you wish to save the <∗.mdp> file before pro-
ceeding as shown in Figure 3.17.
DR
Figure 3.16: Exiting the program
Figure 3.17: Window to save <∗.mdp> file before quitting
After quitting PART-GUI, you are returned to the Particle tracking window. You can then exit
Delft3D entirely by selecting Return and next Exit.
22 of 137 Deltares
4 Graphical User Interface
4.1 General
4.1.1 Initial conditions

D-WAQ PART assumes that there are initially no particles present in the computational do-
main, except for the oil module where initial oil patches can be specified.
T
4.1.2 Boundary conditions
There are no boundary conditions. Particles which are transported out of the model domain
cannot return.
AF
4.1.3 Co-ordinate system
Co-ordinates must be given in the Cartesian (metric) system.
4.1.4 Data groups

The data for a particle tracking scenario are grouped into the following data groups: De-
scription, Hydrodynamics, Substances, Time Frame, Initial Condition, Numerical Parameters,
Instantaneous Releases, Continuous Releases, Process Parameters, Observation Points and
Output Options. Each data group is described in detail in the following sections.
The GUI has been set up in such a way that normally you should work through the data groups
DR
from top to bottom.
4.2 Description
Selecting the Data Group Description opens a window with edit fields as shown in Figure 4.1.
This allows you to add or change text in the input file that will assist in identifying and describ-
ing the scenario.
Domain:
Parameter Lower limit Upper limit Default Unit
Line 1 0 characters 40 characters blank
Figure 4.1: Data Group Description
Deltares 23 of 137
T
AF Figure 4.2: Data Group Hydrodynamics
4.3 Hydrodynamics
Selecting the Data Group Hydrodynamics opens a window as shown in Figure 4.2.
A particle tracking scenario requires the results from an existing hydrodynamic scenario. The
files required for D-WAQ PART are extracted from a Delft3D-FLOW result file by the coupling
module. The coupling module creates a <∗.hyd> file that contains information about the
DR
hydrodynamic scenario and the associated hydrodynamic files (flows, volumes and grid).
To open hydrodynamic results, click Select File. This opens a standard navigation window for
file selection as shown in Figure 4.3. After selecting the required <∗.hyd> file, the names of
the relevant hydrodynamic files appear in the window as shown in Figure 4.2. These names
can only be changed by selecting a different hydrodynamic scenario file.
A hydrodynamic scenario file may be located in another directory other than the working
directory.
4.3.1 Number of layers in hydrodynamics

The number of layers given in the hydrodynamic scenario file will determine whether the
D-WAQ PART simulation will be carried out as a 3-dimensional (3D) simulation or as a 2-
dimensional, depth-averaged (2D) simulation. If the hydrodynamic results are from a 2D
hydrodynamic simulation then only one layer will be present in the hydrodynamic scenario file
and D-WAQ PART will also be a 2D simulation.
If the hydrodynamic results are from a 3D hydrodynamic simulation, the D-WAQ PART simu-
lation can either be 2D or 3D, depending on the number of layers aggregated by you in the
Coupling module.
24 of 137 Deltares
Graphical User Interface
T
AF
Figure 4.3: Navigation window to select a hydrodynamic scenario
DR
Figure 4.4: Data Group Substances when Tracers are selected
Deltares 25 of 137
4.4 Substances
Selecting the Data Group Substances opens a window as shown in Figure 4.4.
Before defining a specific substance in D-WAQ PART, a Type of model must be selected. You
must select either Tracer or Oil.
⋄ Tracer : conservative substances (e.g. salt, dyes), bacteria or other decaying and settling
substances (see Figure 4.4).
⋄ Oil: for modelling the spreading and fate of oil spills (see Figure 4.6).
T
The default option is Tracer.
Only one Type of model can be selected for a D-WAQ PART simulation. However, it is possible
to simulate oil together with a tracer by selecting model type Oil and then adding a specific
AF tracer substance with the tick at Oil fraction removed. A tracer defined in this way (thus in
combination with an oil substance) can only be released at the water surface in the same way
as an oil spill.
4.4.1 Tracers
For the selected model type Tracer, a default substance is given automatically with the name
“tracer” as shown in Figure 4.4.
To add a substance:
⋄ Click Add; a substance named (new) is created.
DR
⋄ Type the name of the substance in the name box and confirm by clicking outside the name
box, but inside the GUI. The name must be a non-blank name.
To rename a substance:
⋄ Select the substance in the list.
⋄ Edit the name of the substance in the name box and confirm by clicking outside the name
box, but inside the GUI. The name must be a non-blank name.
To delete a substance:
⋄ Select the substance in the list.
⋄ Click Delete.
⋄ Confirm that the substance and its related data are to be deleted by selecting Yes in the
confirmation window as shown in Figure 4.5.
⋄ The substance is then erased from the list.
D-WAQ PART requires that at least one substance must be defined.
For Tracer calculations, the units of mass are given in [kg] and the units of concentration in
[kg/m3 ].
Domain:
Name of substance 1 character 14 characters tracer

(non-blank)
26 of 137 Deltares
Figure 4.5: Confirmation window to delete substance
T
AF
DR
Figure 4.6: Data Group Substances when Type of model Oil is selected
4.4.2 Oil
If the Type of model is selected as Oil, the Substances window changes as shown in Fig-
ure 4.6.
The default substance name is given as “Ekofisk”. An additional check box appears to set the
substance to oil or to a tracer. Default the Oil fraction check box is ticked off. If the tick is
removed, the substance will behave as a tracer. However, the tracer in this case can only be
released at the water surface in the same way as an oil spill.
Oil substances can be added, renamed or deleted in the same way as described earlier for
Tracer.
For an oil fraction, the units of mass are given in [kg]. For the computations, D-WAQ PART
assumes that an oil fraction consists of three phases. The units of concentration of an oil
fraction depend on the oil phase:
⋄ Floating Oil; a thin layer of oil floating on top of the water column; [kg/m2 ]
⋄ Dispersed (Entrained) Oil; oil droplets entrained in the water column; [kg/m3 ]
⋄ Sticking Oil; oil that sticks to the land or the bed and is no longer transported; [kg/m2 ]
When an oil fraction is initially released, it is assumed to consist entirely of floating oil. During
the computation, floating oil may be converted to dispersed (entrained) oil or to sticking oil.
The conversion process is controlled by parameters that must be defined under the Data
Deltares 27 of 137
T
AF Figure 4.7: Time Frame Data Group
Group Process Parameters.
Domain:
Name of substance 1 character 14 characters Ekofisk

(non-blank)
DR
Oil fraction check box for Oil fraction or not Oil fraction
an Oil fraction
Restrictions:
⋄ PART-GUI will allow the definition of a maximum of 90 substances. Each oil type rep-
resents 3 simulated substances (i.e. floating, entrained and sticking oil). It is therefore
possible to define a maximum of 30 oil types.
⋄ If you want to use not an Oil fraction substances together with Oil fraction substances,
then you have to define first ALL Oil fraction substances.
4.5 Time frame

Selecting the Data Group Time Frame opens a window as shown in Figure 4.7.
In this data group, you must specify:

⋄ the simulation stop time
⋄ the time step
4.5.1 Date-time format

All times are given in the date-time format “dd mm yyyy hh mm ss” in which:
⋄ dd mm yyyy: date in day, month, year format
⋄ hh mm ss: time of day in hours, minutes, seconds format.
28 of 137 Deltares
Figure 4.8: Data Group Initial Condition
T
4.5.2 Start time
The simulation start time is equal to the start time of the hydrodynamic scenario (after cou-
pling). The value is read from the hydrodynamic result file and cannot be changed.
AF
4.5.3 Stop time
The simulation stop time given in the Stop time field must be greater than the simulation
start time value. If the simulation stop time is greater than the stop time of the hydrodynamic
scenario (after coupling), the hydrodynamic results are repeated as if the flow was periodic.
In this case, you should ensure that the hydrodynamic results have been stored (or coupled)
over a cyclic period and that there is no major discontinuity between the hydrodynamic results
in the first time step and the last time step of the cycle.
DR
4.5.4 Time step

The time step given in the Time step field for the particle tracking computation must be less
than or equal to the time step in the hydrodynamic scenario result file. Furthermore, to avoid
interpolation errors, the particle tracking time step must be chosen such that it is an integer
fraction of the hydrodynamic scenario time step.
4.5.5 Reference time input

For the simulation, all time instances and time-series are expressed in the number of time
steps relative to the reference time input. To ensure consistency, the reference time input for
the particle tracking simulation must be equal to the reference time input of the hydrodynamic
scenario. This value is read from the hydrodynamics result file by the coupling module and
cannot be changed.
4.6 Initial condition

Initial conditions can only be specified when the oil module is selected. Then the Initial Con-
dition Data Group will be available. Selection of this data group opens a window as shown in
Figure 4.8.
If you select to specify the initial condition, then a navigation window will open that enables
you to select the initial condition file (<∗.ini>). See section A.6 for details of the <∗.ini> file.
It should be noted that the number of particles specified in the initial condition file is in addition
to the number of particles associated with instantaneous and continuous releases.
Deltares 29 of 137
T
Figure 4.9: Data Group Numerical Parameters
AF
4.7 Numerical parameters
Selecting the Numerical Parameters Data Group opens a window as shown in Figure 4.9.
In this data group you must specify:

⋄ the number of particles per release
⋄ a vertical dispersion option
⋄ a scaling factor for the vertical dispersion coefficient
DR
4.7.1 Number of particles

The number of particles per release used in the simulation is important for the accuracy of
the computed results. Due to the stochastic nature of the dispersion process, the simulation
requires a large number of particles to obtain the best results. In general, the accuracy varies
with the square-root of the number of particles.
The number of particles required will depend on several factors, namely:

⋄ the length of the simulation
⋄ the expected extent of spreading of the particles
⋄ the size of the cells in the computational grid or the zoom window
⋄ the number of releases
⋄ the number of layers used in vertical direction, and the accuracy required over vertical
direction
Furthermore, continuous releases usually require more particles than instantaneous releases.
The accuracy of a simulation is often expressed as the smallest concentration that can be
represented by a single particle in a computational cell. The minimum concentration level that
can be represented by a particle is the mass of the particle divided by the volume of the cell
in which the particle is located.
mparticle Mtotal
Cmin = = (4.1)
Acell × hlayer Ntotal × Acell × hlayer
where:
Cmin minimum concentration resolution [kg/m3 ]
30 of 137 Deltares
mparticle mass per particle [kg]

Acell surface area of computational (or zoom window) cell [m2 ]
hlayer local layer thickness [m]
Ntotal total number of particles in the release [-]
Mtotal total mass of release [kg]
Example
The above relationship can be used to estimate the required number of particles for a given
simulation. Consider an example in which you wish to simulate the release of a total of
T
10 000 kg of substance. The smallest grid cell in the computational domain within the area of
interest has a surface area of 200 m2 with a minimum layer thickness of 4 m. It is required that
the simulation should be able to resolve concentrations of at least 0.001 kg/m3 . The minimum
total number of particles required for this release would therefore be N ≥ 10 000/(0.001 ·
200 · 4) = 12 500 particles.
AF If you require even more resolution of the concentration level (i.e. even smaller minimum
concentration level) then the mass per particle must be reduced. This is usually done by
increasing the number of particles for a given release. In practice the best compromise be-
tween accuracy (e.g. resolution) and performance (e.g. speed of computation) is usually found
by trial and error.
4.7.2 Vertical dispersion

You must select a vertical dispersion option by clicking on the appropriate radio button. The
DR
options are:
⋄ Depth averaged algebraic: In this option, the vertical dispersion is assumed to be constant
over the depth. The value is approximated by an analytical function described in Chapter
6. The analytical function calculates the vertical dispersion coefficient as a function of
the total depth. In this case, therefore, the vertical dispersion coefficient may vary with
location if the total depth is also varying.
⋄ Constant dispersion: In this option, the vertical dispersion coefficient is taken to be con-
stant in time and space. In this case, you must specify a (constant) value for the vertical
dispersion coefficient.
A vertical dispersion Scale factor must be given in the edit field. The vertical dispersion
coefficient defined above is multiplied by this scale factor for the simulation. The default value
is 1.0.
Deltares 31 of 137
T
Figure 4.10: Instantaneous Releases Data Group for tracer substances
AF Domain:
Number of particles 1 per unlimited2) 10 000 -

substance1)
Vertical dispersion option Depth averaged algebraic Constant -

Constant
Dispersion coefficient 0 - 0 m2 /s
DR
Scale factor 0.0 1.0 1.0 -
1)
It is not recommended to use less than 100 particles per substance.
2)
If Particle tracks is selected in the Output Options Data Group, the number of particles should be
less than 1000.
4.8 Instantaneous releases

In the Data Group Instantaneous Releases you can specify the name and co-ordinates of the
instantaneous release location, the initial radius of the release and the number of particles to
be released. The mass of substance released at each release location and the release time
are defined in the Release tables for instantaneous releases window.
Selecting the Data Group Instantaneous Releases opens a window as shown in Figure 4.10.
This window belongs to tracer substances. For oil substances the window is slightly different,
see further on.
To add an instantaneous release location:
⋄ Press Add; an instantaneous release named (new) is created.

⋄ Type the name of the release location in the name box and confirm by pressing Enter on
the keyboard. The name must be a non-blank name.
⋄ Use the Visualisation Area window or the keyboard to specify the x co-ordinate and the
y co-ordinate of the instantaneous release location.
⋄ Specify the vertical position of the instantaneous release location in one of two ways:
a for 2D models; as a percentage of the local water depth (0 % = surface; 100 % =
32 of 137 Deltares
T
Figure 4.11: Release tables for instantaneous releases window for defining release
AF time and mass of release
bottom);
b for 3D models; as a layer number in which the release is uniformly distributed over the
entire layer (layer 1 = surface layer)
⋄ Specify the number of particles to be used for this release.
⋄ Specify the radius of the instantaneous release location.
⋄ Press Release Tables to open the sub-window shown in Figure 4.11.
⋄ Specify the release time.
⋄ Specify the mass of each of the substances to be released as defined in the Data Group
Substances.
DR
To delete an instantaneous release location:
⋄ Select the instantaneous release location in the list.

⋄ Click Delete.
⋄ Confirm that the release and its related data are to be deleted by selecting Yes in the
confirmation window.
⋄ The instantaneous release is then deleted from the list.
4.8.1 Co-ordinates of instantaneous release location

The horizontal position of the instantaneous release location is given in Cartesian co-ordinates
defined by an x co-ordinate and a y co-ordinate. The co-ordinates of the instantaneous
release location must be within the computational grid.
The definition of the vertical position of the instantaneous release location will depend on
whether a 2D or 3D simulation is being defined.
⋄ For 2D simulations, there is only one (horizontal) layer in the model and the vertical posi-
tion of the instantaneous release location must be defined as a percentage of the depth
below the water surface. That is, 0 % represents a surface release and 100 % represents
a release at the bottom of the water column.
⋄ For 3D simulations, the released particles are distributed uniformly over the layer in which
the instantaneous release is defined. You must therefore give the layer number in which
the release takes place. Layer 1 represents the surface layer and is the default value.
For oil releases, the substance to be released is floating oil and therefore the particles can
only be released at the surface. For 2D models, the floating oil must be released at 0 % of the
Deltares 33 of 137
T
AF
Figure 4.12: Instantaneous Releases Data Group for oil substances
depth below the surface. For 3D models, the floating oil must be released in Layer 1 (surface
layer). Therefore, if the oil module has been selected in the Data Group Substances, the
vertical position of the instantaneous release location is fixed and cannot be altered by you,
see Figure 5.13.
DR
4.8.2 Particles used

You must define how many particles are to be used for each instantaneous release. The
number of particles used per release will depend on the required resolution of the computa-
tion. The number of particles used for each release is defined as a percentage of the number
of particles given in the Data Group Numerical Parameters. For example, if the number of
particles per release was defined as 10 000 in the Numerical Parameters Data Group and
you have defined two releases each using 100 % of the particles, then the total number of
particles that will be used in the simulation will be 2 × 10 000 = 20 000 particles.
4.8.3 Radius of instantaneous release

D-WAQ PART can be used for studying the spread of substances in the mid-field region of
interest. D-WAQ PART does not model the initial spreading and mixing (i.e. near-field) phe-
nomena that occur as a discharge leaves a nozzle or outlet pipe and enters the ambient wa-
ters. The initial spreading and mixing occurs very quickly and quite close to the outlet point. In
order to incorporate the effect of initial spreading in a D-WAQ PART simulation, it is possible
to define a radius of the instantaneous release. The particles in the instantaneous release
are then spread uniformly around the circumference of a circle with this radius, centred at the
release location.
For oil spills, the initial radius of the instantaneous release may be determined by the gravi-
tational surface spreading that occurs very quickly after oil is released at the water surface,
see Chapter 6. If the oil module has been selected in the Substances Data Group, it is pos-
sible to allow D-WAQ PART to calculate this initial radius by selecting the appropriate option
in the Data Group Instantaneous Releases. The radius is then calculated from a user-defined
density of oil [kg/m3 ]. You can also specify a user-defined radius, see Figure 4.12.
34 of 137 Deltares
For the oil module it is recommended to use the calculated radius because this will allow the
specification of the oil density. If oil density is not provided a value of 890 kg/m3 is assumed.
4.8.4 Release time and mass of release

The release time and the mass of the instantaneous release are defined in the Release tables
for instantaneous releases window as shown in Figure 4.11. The (instantaneous) Release
time must be given in date-time format. The Substances table contains one column for each
substance defined in the Data Group Substances. You must specify the total mass of each of
the substances released in kg.
T
Domain:
Name of instantaneous re- 1 character 20 characters (new) -

AF lease
x co-ordinate - - 0 m
y co-ordinate - - 0 m
Layer number (for 3D only) 1 no. of layers 1 -
Percentage of depth below sur- 0 (surface) 100 (bottom) 0 %

face (for 2D only)
Particles used 0 100 100 %

DR
Radius of release option user-defined user-defined

calculated (only for oil)
Radius of release 0 - 0 m
Release time start time stop time start time date-

time
format
Amount of substance released 0 - 0 kg
4.9 Continuous releases

In the Data Group Continuous Releases you can specify the name and co-ordinates of the
continuous release location, the initial radius of the release and the number of particles to be
released. The discharge rate of the release, the concentration of the substance released and
the release time are defined in the Release tables for continuous releases window.
Selecting the Data Group Continuous Releases opens a window as shown in Figure 4.13.
This window belongs to tracer substances. For oil substances the window is slightly different,
see further on.
To add a continuous release location:
⋄ Click Add; a continuous release named (new) is created.

⋄ Type the name of the location in the name box and confirm by pressing Enter on the
keyboard. The name must be a non-blank name.
⋄ Use the Visualisation Area window or the keyboard to specify the x co-ordinate and the
Deltares 35 of 137
T
Figure 4.13: Continuous Releases Data Group for tracer substances
AF y co-ordinate of the continuous release location.
⋄ Specify the vertical position of the continuous release location in one of two ways:
a for 2D models; as a percentage of the local water depth (0 % = surface; 100 % =
bottom);
b for 3D models; as a layer number in which the release is uniformly distributed over the
entire layer (layer 1 = surface layer)
⋄ Specify the number of particles to be used for this release.
⋄ Specify the radius of the continuous release.
DR
⋄ Click Release Tables to open the sub-window shown in Figure 4.14.

⋄ Specify the time-series of discharge for each release.
⋄ Select a radio button to indicate whether Linear or Block interpolation is to be used in the
time-series table.
⋄ Specify the concentration of each of the substances in the continuous release as defined
in the Substances Data Group.
To delete a continuous release location:

⋄ Select the continuous release location in the list.
⋄ Click Delete.
⋄ Confirm that the release and its related data are to be deleted by selecting Yes in the
confirmation window.
⋄ The continuous release is then deleted from the list.
4.9.1 Co-ordinates of continuous release location

The horizontal position of the continuous release location is given in Cartesian co-ordinates
defined by an x co-ordinate and a y co-ordinate. The co-ordinates of the continuous release
location must be within the computational grid.
The definition of the vertical position of the continuous release location will depend on whether
a 2D or 3D simulation is being defined.
⋄ For 2D simulations, there is only one (horizontal) layer in the model and the vertical po-
sition of the continuous release location must be defined as a percentage of the depth
below the water surface. That is, 0 % represents a surface release and 100 % represents
a release at the bottom of the water column.
36 of 137 Deltares
T
AF Figure 4.14: Release tables for continuous releases window for defining release time,
discharge and concentration of release
⋄ For 3D simulations, the released particles are distributed uniformly over the layer in which
the continuous release is defined. You must therefore give the layer number in which the
release takes place. Layer 1 represents the surface layer and is the default value.
DR
For oil releases, the substance to be released is floating oil and therefore the particles can
only be released at the surface. For 2D models, the floating oil must be released at 0 % of the
depth below the surface. For 3D models, the floating oil must be released in Layer 1 (surface
layer). Therefore, if the oil module has been selected in the Data Group Substances, the
vertical position of the continuous release location is fixed and cannot be altered by you.
4.9.2 Particles used

You must define how many particles are to be used for each continuous release. The number
of particles used per release will depend on the required resolution of the computation. The
number of particles used for each release is defined as a percentage of the number of particles
given in the Data Group Numerical Parameters. For example, if the number of particles per
release was defined as 10 000 in the Numerical Parameters Data Group and you have defined
two releases each using 100 % of the particles, then the total number of particles that will be
used in the simulation will be 2 × 10 000 = 20 000 particles.
4.9.3 Radius of continuous release

D-WAQ PART can be used for studying the spread of substances in the mid-field region of
interest. D-WAQ PART does not model the initial spreading and mixing (i.e. near-field) phe-
nomena that occur as a discharge leaves a nozzle or outlet pipe and enters the ambient
waters. The initial spreading and mixing occurs very quickly and quite close to the outlet
point. In order to incorporate the effect of initial spreading in a D-WAQ PART simulation, it is
possible to define a radius of the continuous release. The particles in the continuous release
are then spread uniformly around the circumference of a circle with this radius, centred at the
release point.
Deltares 37 of 137
4.9.4 Release time and rate

The release rate per substance in m3 /s is defined in the time-series table in the Release rates
section of the Release tables for continuous releases window as shown in Figure 4.14. The
radio buttons alongside the table allows you to specify the Interpolation method that is to be
used to define the time-series functions. The time-series function can be defined in one of the
following two ways:
⋄ Linear interpolation method

The time-series of discharges is defined as a piecewise linear time-series. A series of dis-
charges [m3 /s] are given at various time breakpoints (in time-date format). D-WAQ PART
T
interpolates linearly between these time records. At least two time breakpoints must be
defined.
⋄ Block interpolation method
The time-series of discharges is defined as a series of block functions. A series of dis-
charges [m3 /s] are given at various time breakpoints (in time-date format). D-WAQ PART
AF uses the discharge associated with a given time breakpoint as a constant value until a
new time breakpoint and associated discharge is defined.
Remark:
⋄ The Generate table button can be used to define a time-series with a constant user-
defined time interval.
A Scale factor can be defined that is applied to all discharge values in the table. For example,
if a time-series has been defined in litres per second [l/s], this can be converted to m3 /s by
using a scale factor of 0.001.
DR
You must define release rates for the entire simulation period. That is, the first time breakpoint
must be equal to the simulation start time and the last time breakpoint must be equal to the
simulation stop time.
4.9.5 Concentration of continuous release

The mass rate of release of a substance [kg/s] is given by the product of its discharge [m3 /s]
and the concentration of the substance in the discharge [kg/m3 ]. The concentration of each
substance in the continuous release is defined in the Concentrations table in the Release
tables for continuous releases window as shown in Figure 4.14.
In order to avoid a situation in which a given particle is used to represent several different
substances at the same time, it is not recommended to define more than one non-zero con-
centration of substance per release location. If several substances are to be released at
the same location, you should define separate release locations for each substance at this
location.
38 of 137 Deltares
Figure 4.15: Process parameters Data Group
Figure 4.16: Process Parameters Data Group for oil modelling
T
Domain:
AF Name of continuous release
x co-ordinate
1 character
-
20 characters
-
(new)
0
-
y co-ordinate - - 0 m
Layer number (for 3D only) 1 no. of layers 1 -
Percentage of depth below sur- 0 (surface) 100 (bottom) 0 %

face (for 2D only)
Particles used 0 100 100 %

DR
Radius of release 0 - 0 m
Interpolation method block or linear linear -
Scale factor 0 - 1 -
Release rate: time breakpoints start time stop time start time date-
time
format
Release rate: discharge 0 - 0 m3 /s
Concentrations 0 - 0 kg/m3
4.10 Process parameters

Selecting the Data Group Process Parameters opens a window as shown in Figure 4.15. If
Oil has been selected in the Data Group Substances, the Process Parameters Data Group
window will include an extra Oil Model check button as shown in Figure 4.16.
In this Process Parameters Data Group you can specify:
⋄ decay rates
⋄ sedimentation/erosion parameters
⋄ physical parameters
⋄ oil-specific parameters, if applicable.
Deltares 39 of 137
T
AF
Figure 4.17: Decay rates sub-window
4.10.1 Decay rates

Selecting the Sub-data Group Decay Rates opens the Decay Rates sub-window as shown in
Figure 4.17.
DR
The time-dependent decay rate for each of the substances defined in the Data Group Sub-
stances must be specified in the table. The time-series of decay rates is defined as a piece-
wise linear time-series. A series of decay rates [1/day] are given at various time breakpoints
(in time-date format). D-WAQ PART interpolates linearly between these time records. At least
two time breakpoints must be defined covering the entire simulation period. That is, the first
time breakpoint must be equal to the simulation start time and the last time breakpoint must
be equal to the simulation stop time.
Remarks:
⋄ The Table option in the menu bar can be used for the manipulation of table data.
For conservative tracers the decay rate is zero. If Oil has been selected in the Data Group
Substances, the decay rate given here will be in addition to the model-specific decay pro-
cesses (e.g. entrainment, evaporation, emulsification, etc.) that are defined in the Oil Model
sub-window.
Domain:
Decay rates: time breakpoints start time stop time start time date-
time
format
Decay rates: decay rate 0 - 0 1/day
40 of 137 Deltares
Figure 4.18: Sedimentation/erosion sub-window
T
AF
Figure 4.19: Sediment/Erosion constants sub-window
DR
4.10.2 Sedimentation/erosion
Selecting the Sub-data Group Sedimentation/erosion opens the sub-window with two check
buttons as shown in Figure 4.18.
Sediment/erosion constants
Selecting Constants will open the sub-window shown in Figure 4.19.
You must specify whether to Include or Exclude sediment/erosion processes in the computa-
tion by selecting the appropriate radio button.
If sediment/erosion processes are to be included, additional parameters must be defined:

⋄ Chézy coefficient [m1/2 /s]
⋄ critical shear stress for sedimentation [Pa]
⋄ critical shear stress for erosion [Pa]
If the bed shear stress at any location is less than the critical shear stress for sedimentation,
a particle that comes into contact with the bottom at that location, will remain attached to the
bottom (sedimentation). For sedimentation, D-WAQ PART creates an extra model layer for
sediment at the bed.
If the bed shear stress at any location is greater than the critical shear stress for sedimentation,
a particle that comes in contact with the bottom at that location will be reflected back into the
water column.
If the bed shear stress at any location is greater than the critical shear stress for erosion, all
Deltares 41 of 137
T
AF
Figure 4.20: Settling velocities sub-window
deposited particles at that location (i.e. particles located in the extra bed-sediment layer) will
be returned to the water column instantaneously.
Domain:
DR
Sediment/erosion processes include or exclude exclude -
Chézy coefficient 0 - 55 m1/2 /s
Critical shear stress sedimen- 0 - 0.05 Pa

tation
Critical shear stress erosion 0 - 0.4 Pa
Settling velocities
Selecting Settling velocities will open the sub-window shown in Figure 4.20.
You must define the coefficients A0 and A1 , the period T and the phase φ as piecewise
linear functions of time by selecting the appropriate check button in the Settling velocities
sub-menu. In addition, you must also define the minimum and maximum allowable values
for the calculated settling velocity by selecting the Vmin and Vmax check buttons respectively.
Vmin and Vmax must also be defined as piecewise linear functions of time.
Values for all of the above parameters must be defined at various time breakpoints (in time-
date format). D-WAQ PART interpolates linearly between these time records. At least two
time breakpoints must be defined. You must define parameter values for the entire simulation
period. That is, the first time breakpoint must be equal to the simulation start time and the last
time breakpoint must be equal to the simulation stop time. The Table option in the menu bar
can also be used to assist in the creation and manipulation of the table data.
Remarks:
42 of 137 Deltares
⋄ The time breakpoints must be the same for A0 , A1 , T , φ and Vmin and Vmax .
⋄ If the calculated settling velocity lies outside the interval [Vmin , Vmax ] then the value is
reset to Vmin if the calculated value is less than Vmin . Similar for Vmax .
Domain:
Exponent for concentration de- 0 - 0 -

pendent settling velocity
grid refinement factor 0 - 1 -
T
A0: time breakpoints start time stop time start time date-
time
format
A0: value 0 - 0 m/s

AF A1: time breakpoints start time stop time start time date-
time
format
A1: value 0 - 0 m/s
Period: time breakpoints start time stop time start time date-
time
format
Period: value 0 - 0 hours

DR
Phase: time breakpoints start time stop time start time date-
time
format
Phase: value 0 - 0 hours
Vmin: time breakpoints start time stop time start time date-
time
format
Vmin: value 0 - 0 m/s
Vmax: time breakpoints start time stop time start time date-
time
format
Vmax: value 0 - 1000 m/s
4.10.3 Physical parameters

Selecting the Sub-data Group Physical will open the sub-window as shown in Figure 4.21.
This window is used to define the parameters related to the horizontal dispersion coefficient
and the wind speed and direction.
Horizontal dispersion coefficient

The time-dependent horizontal dispersion coefficient is given as in Chapter 6.
Deltares 43 of 137
T
AF
Figure 4.21: Physical parameters sub-window
Roughness length (2DH only)

If the hydrodynamic results from Delft3D-FLOW are 2-dimensional (2DH), D-WAQ PART will
superimpose a logarithmic velocity profile in the vertical direction to emulate 3-dimensional
hydrodynamic effects (see Chapter 6). This is the so-called 2.5 DH mode. The roughness
DR
length [m] defined here is a calibration parameter used to calculate the logarithmic velocity
profile based on bottom shear stress.
Effect of wind
For 2DH hydrodynamics, the horizontal wind velocity given here is used to superimpose a
vertical velocity profile over the depth-averaged velocity value from the hydrodynamic result
file (see Chapter 6).
For 3D hydrodynamics, the horizontal wind velocity given here is only used if Oil has been
selected in the Data Group Substances. In this case, the horizontal wind velocity is used to
determine the additional wind-induced drift/deflection on surface floating oil (see Chapter 6).
The horizontal wind velocity is defined by:
⋄ wind drag coefficient, typically 1 %–3 %.

⋄ wind speed at 10 m above the sea level
⋄ wind direction in degrees relative to North
Remarks:
⋄ Wind direction is given as the direction from which the wind is blowing. For example,
wind from the North is 0.0 degrees.
⋄ When the wind direction passes the North from one time point to another, the interpo-
lation will fail. For instance: 350 degrees at time point T1 and 10 degrees at time point
T2 , will give 180 for the time point in the middle. To overcome this specify 370 degrees
in stead of 10 degrees.
Both wind speed and direction must be specified as a piece-wise linear function of time. Wind
effects can be applied on all tidal events and are in addition to the (overall) wind effects applied
44 of 137 Deltares
in the hydrodynamic scenario.
Remarks:
⋄ The Table option in the menu bar can be used for the manipulation of table data.
Density of water
⋄ density of water [kg/m3 ]
T
Domain:
Horizontal dispersion: 0 - 1 m2 /s
AF coefficient a
Horizontal dispersion:
coefficient b
0 - 0.01 -
Roughness length (2DH only) 0 - 0.02 m
wind drag 0 - 1 %
Density of water 0 - 1024 kg/m3
wind: time breakpoints start time stop time start time date-
DR
time
format
wind: wind speed 0 - 0 m/s
wind: wind direction 0 360 0 degrees
4.10.4 Oil Model parameters

Oil Model parameters must be supplied if Oil is selected in the Substances Data Group.
Selecting the Sub-data Group Oil Model will open the sub-window as shown in Figure 4.22.
The oil spill model requires input to describe the following physical processes:
⋄ evaporation
⋄ beaching (sticking probability)
⋄ emulsification
⋄ dispersion (entrainment)
⋄ decay
⋄ initial spreading
⋄ deflection due to Coriolis
Additional input fields for defining dispersion (entrainment), sticking probability and other oil
parameters are available.
Deltares 45 of 137
T
AF
Figure 4.22: Oil Model input for the Oil model
Evaporation
DR
Evaporation in the oil model is described as a first-order decay process. The evaporation rate
for each oil type must be defined by a decay coefficient, kE where:
dm
= −kE t (4.2)
dt
where kE is the evaporation rate in 1/day and t is the time in days.
The rate may be specified as a constant or be determined according to Fingas (see Section
refoilmodel). In the latter case two extra parameters are to be specified, the mass percentage
of the oil that evaporates at 180o C and the water temperature. (NOte: The label to the fraction
is currently not updated.)
Sticking probability (beaching)

In order to simulate the process of oil sticking to land or to the bed of the water column,
a sticking probability must be specified. A particle may come into contact with land due to
wind effects or horizontal dispersion and it may come into contact with the bed due to vertical
dispersion or settling. The particle sticks to the land or bed if a randomly chosen number
between 0 and 1 is smaller than the sticking probability given. For a sticking probability value
of 1.0, all particles that come into contact with land or the bed will stick and become inactive
particles (i.e. they will not be transported by water anymore). These particles are referred to
as Sticking oil in the D-WAQ PART output.
Volatile fraction
This parameter, with a range of 0–1, specifies the fraction of the oil that can evaporate. The
remainder will not evaporate.
46 of 137 Deltares
Emulsification parameter
If this parameter is set to 0, then emulsification is switched off. When you decide to switch
emulsification on, then it is recommended to use the default value of 2 · 10−6 .
Maximum water content

The maximum water content of the emulsion, ranging from 0–1, is required. It is an oil type
dependent parameter.
T
Fraction at which emulsification starts
This parameter is dependent on the oil composition and may vary between 0–1. If emulsifica-
tion occurs then a value of the order of 0.05–0.10 may be expected.
AF
Density of oil
The density of oil is expressed in kg/m3 .
Kinematic viscosity of oil

The viscosity of oil is its resistance to shear or flow and is a measure of the oil adhe-
sive/cohesive or frictional properties.
Dispersion (entrainment)
DR
The entrainment of oil in water is referred to as dispersion. The dispersion rate may be defined
using one of two dispersion options.
User-defined dispersion
The dispersion (entrainment) process is described by a first order process. The dispersion
rate for each oil type must be defined by a decay coefficient, kD where:
dm
= −kD t (4.3)
dt
where kD is the dispersion rate in 1/day and t is the time in days.
Calculated dispersion
The dispersion rate for each oil type is calculated directly by D-WAQ PART using the
formulae given in Chapter 6. In this case, the density of oil should be supplied by you. By
default the wind velocity at which white capping is supposed to occur is set to 5 m/s, but it
is possible to change that.
Minimum oil layer thickness

This parameter is used in the calculation of the dispersion (entrainment). Validation of the
model has shown that the default value of 0.0005 m generally gives good dispersion rates.
The parameter may be used to calibrate the dispersion process, if data is available. Otherwise
it is recommended to use the default value.
Deflection angle (Coriolis effect, 3D only)

The advection of surface floating oil is subject to additional wind effects. The currents induced
by wind and waves may act in addition to the currents described by the overall (e.g. tidal)
hydrodynamics. The deflection angle is the angle between the wind direction and the actual
direction of advection of the floating oil. The difference between the two directions arises
Deltares 47 of 137
because the waves move under the influence of Coriolis so that wave-induced transport is at
an angle to the wind.
The deflection angle (nautical convention) can be defined in the data field as shown in Fig-
ure 4.22. The wind velocity and the drag coefficient are defined in the Physical sub-menu of
the Data Group Process Parameters.
Remark:
⋄ The deflection angle is not used for 2DH hydrodynamics. For 2DH hydrodynamics,
a wind-induced parabolic velocity profile is superimposed over the entire depth. The
T
deflection angle at the surface is not applicable here.
Domain:
AF
Evaporation rate
Sticking probability
0
0
-
1
0
0.5
1/day
Volatile fraction 0 1 0.94 -
Emulsification parameter 0 - 0 -
2 · 10−6 1)
Maximum water content 0 1 1 -

0.7 1)
DR
Fraction at which emulsifica- 0 1 0 -

tion starts 1 1)
Density of oil 0 - 0890 kg/m3

(to calculate dispersion rate)
Kinematic viscosity 0 - 0 cSt

1)
(to calculate dispersion rate) 1500
Dispersion option user-defined or calcu- calculated -

lated
User-defined dispersion rate 0 - 0 1/day
Minimum oil layer thickness 0 - 0.0005 m
Deflection angle -360 360 0 degrees
1)
These values are default values, when importing an input file generated from previous PART-GUI
releases.
48 of 137 Deltares
T
Figure 4.23: Observation Points Data Group
AF
4.11 Observation points
Computational results can be monitored as a function of time by using observation points.
Results for the observation points are written to a history file at user-specified time intervals.
The time intervals for the output are defined in the Data Group Output Options. At the obser-
vation points, all computed quantities are stored as a function of time for post-processing and
visualisation.
Selecting the Data Group Observation Points opens a window as shown in Figure 4.23.
DR
Observation points are characterised by a name, and the co-ordinates of its location. Co-
ordinates must be given in the Cartesian system.
To add an observation point:

⋄ Click Add; an observation point named (new) is created.
⋄ Type the name of the observation point in the name box and confirm by pressing Enter on
the keyboard. The name must be a non-blank name.
⋄ Use the Visualisation Area window or keyboard to specify the x co-ordinate and the y
co-ordinate of the observation point.
To delete an observation point:

⋄ Select the required observation point from the list.
⋄ Click Delete.
⋄ Confirm that the observation point is to be deleted by selecting Yes in the confirmation
window.
⋄ The observation point is then deleted from the list.
Observation points must be within the computational model area. In addition, an observation
point should be located at some distance from a model boundary in order to provide reliable
results. Half a grid interval is advised as a minimum distance from a boundary.
Deltares 49 of 137
T
Figure 4.24: Output Options Data Group
AF Use of Zoom grid
The values of concentration that are output at the observation points are the cell-averaged
concentration values. Computational cells are generally quite large so that two observation
points that are located close to each other within the same computational cell will thus give the
same value. The resolution and accuracy of the output concentrations can be increased by
using the so-called Zoom grid, which is defined in the Data Group Output Options. The Zoom
grid allows you to define smaller cell sizes for the output. If an observation point is located
within the Zoom grid, the output concentrations provided are the Zoom grid cell-averaged
concentrations. If the observation point is located outside of the Zoom grid, the output con-
centrations provided are the computational cell-averaged concentrations.
DR
4.12 Output options

The Data Group Output Options allows you to specify which results need to be stored for
visualisation and at which time interval. Selecting the Data Group Output Options opens a
window as shown in Figure 4.24.
Output from D-WAQ PART is stored in several different ways, namely:

⋄ for selected observation points, at specified time intervals (history file)
⋄ for the entire computational grid, at specified time intervals (map file)
⋄ for a user-defined area, known as the Zoom Grid Output at specified time intervals
Output files
For each of the above output options, you must define the start time, the stop time and a time
interval for the outputs. The output start time must be equal to or greater than the start time of
the simulation. The output stop time must be greater than the output start time and the time
interval for storing output must be an integer multiple of the time-step used in the simulation.
History file
The history file contains time-series of the computational results at each of the locations de-
fined in the Data Group Observation Points.
50 of 137 Deltares
T
Figure 4.25: Definition of Zoom grid
4.12.1 Map file

The map file contains the computational results at every computational cell in the domain.
AF The concentrations in the map file are the computational cell-averaged concentration values.
4.12.2 Zoom grid

The use of a zoom grid is compulsory.
The size of the computational cells is often so large that it is not possible to capture steep
concentration gradients in the map file output. The resolution and accuracy of the output
concentrations can be increased by using the so-called Zoom grid. The zoom grid allows you
to define smaller cell sizes for a better presentation of the output results.
DR
The zoom grid does not affect the computation of the processes. However, the following
should be noted in relation to concentration-dependent settling.
If you choose to define a concentration-dependent settling velocity by selecting an exponent

value n ̸= 0, the local concentration of particles [kg/m3 ] used in this formulation is taken
from the zoom grid. The zoom grid should therefore be defined to cover the entire expected
trajectory of the particles during the simulation.
The most important role of the zoom grid is to improve the spatial resolution of the output
results.
Definition of zoom grid

The extent and grid cell size of the zoom grid is defined by selecting Grid Definition in the
Zoom grid sub-window as shown in Figure 4.25. In order to define extent of the zoom grid,
you must define the beginning and the end co-ordinates of the zoom grid. The size of the zoom
grid cells is then determined by defining the number of cells to be used in the x direction and
the number of cells to be used in the y direction.
The extent of the zoom grid can also be defined with the aid of the Visualisation Area window.
In this case, you must first select the check box Definition via Zoom Box (in Visualisation area;
the Visualisation Area window will then open.
In the Visualisation Area window, you must select Zoom → Zoom Box as shown in Fig-
ure 4.26.
The extent of the zoom grid can then be defined graphically using the mouse to click and drag
Deltares 51 of 137
T
AF
Figure 4.26: Defining zoom grid using the Visualisation Area window
DR
Figure 4.27: Definition of Zoom grid output time intervals
the outline of the desired grid extent. The zoom grid is presented in the visualisation window
by solid red boundary lines. As soon as a new zoom grid is specified a new red box appears.
It is only possible to define one zoom grid.
The x and y co-ordinates of the begin and end points of the zoom grid are transferred auto-
matically to the input fields in the Zoom grid sub-window. You can then edit these values if
desired.
Zoom grid output

The concentration values [kg/m3 ] in each zoom grid cell are calculated from the total mass
of particles present in a cell divided by the volume of the cell. You must define the output
times at which the concentrations in the zoom grid cells are to be output. Selecting Zoom
Grid Output in the Zoom grid sub-window allows you to input the desired time intervals in the
time breakpoints edit field as shown in Figure 4.27.
The time intervals must be given in time-date format. In addition, a recovery factor can be
defined at each time interval. The recovery factor is a factor that can be used to relate the
simulation results to field measurements. The results of the simulation are multiplied by this
factor before being written to file. The results can thus be adjusted to match the measure-
52 of 137 Deltares
ments to take into account processes that may not have been included in the particle tracking
simulation (e.g. additional losses due to evaporation or absorption to silt to the bottom).
4.12.3 Particle tracks

If the 3D tracks check box has been selected, D-WAQ PART will create an extra output file that
contains information about the 3D location of every particle in the simulation at every time step.
The size of this file will grow explosively for a large number of particles and for long simulation
times. It is therefore suggested to use only a very limited number of particles when using
this option (e.g. < 1000). The particle tracking output cannot be visualised using GPP. The
T
data can be visualised using Delft3D-QUICKPLOT or alternatively using the Delft3D-MATLAB
interface toolbox in MATLAB®. Please contact Deltares for further information.
4.12.4 Output file types

AF Each of the following file types can be visualised and plotted using the post-processing pro-
grams GPP or QUICKPLOT that are supplied with Delft3D.
Outputs for the history file are written to two files:

⋄ DELWAQ | Binary history file <∗.his>
⋄ Delft3D | Mid field history file <his∗.dat>
These files contain the same information but are written in different file formats.
Outputs for the map file are written to two files:

DR
⋄ DELWAQ | Binary map file <∗.map>

⋄ Delft3D | Mid field map file <map∗.dat>
Remark:
⋄ When plotting from the map files, first create the model grid (in GPP, Delft3D | Water
quality grid file).
Outputs for the Zoom grid output are written to two files:
⋄ DELPAR | Binary plot grid file <∗.plo>
⋄ Delft3D | Mid field plot grid file <plo∗.dat>
The following output files cannot be visualised using GPP. Please contact Deltares for more
information.
Output for the 3D(-Particle) tracks results are written to the file:
⋄ Delft3D | tracking file <trk∗.dat>
For the oil module a <type.csv> file is generated for each oil type with the name [type],
as specified in the input file. It is an ASCII file containing the overall oil budget (floating,
dispersed, evaporated and sticky) and average surface oil density, surface oil viscosity and
water content. This file can be directly important into Excel.
Deltares 53 of 137
4.13 Executing the scenario
4.13.1 Executing the water quality mid-field module D-WAQ PART

The particle tracking program D-WAQ PART can be executed within Delft3D-MENU. The input
files for running D-WAQ PART are created by PART-GUI and contain all of the information
required to execute D-WAQ PART. This means that no further input or manipulations are
required from you after execution of the program is commenced.
Error handling
T
If an error occurs during the execution of a D-WAQ PART scenario, an error message is
written to the ASCII report file <∗.out>. The error message can be found in the report file by
searching for the word ‘ERROR’ in the text. The report file can be viewed by selecting Reports
on the Particle tracking main menu or by using any other ASCII text editor or viewer.
AF Some common errors include:
⋄ using a quotation mark in the ‘description’ input field
⋄ specifying a stop time that is less than the start time
⋄ specifying incorrect stop times for <∗.his> file and <∗.map> file
⋄ specifying a non-increasing series of plot times for results from the zoom grid
⋄ specifying observation points or releases outside of the active grid (e.g. on land)
⋄ specifying a bottom roughness coefficient of zero
4.13.2 Estimating run time

DR
The run-time of D-WAQ PART depends mainly on the number of particles used and the kind of
computer and processor used. For instantaneous releases, the CPU-time increases linearly
with the number of particles. For continuous releases the CPU-time increases quadratically
with the number of particles. For example, the tutorial example <fti_tracer.inp> supplied
with D-WAQ PART, using a total of 100 000 particles requires approximately 2 minutes of
CPU time on a Pentium IV, 1.6 GHz computer running Microsoft Windows. However, the
same example run using a total of 1 million particles requires approximately 15 minutes on
the same computer. Frequent outputs from the zoom grid and large numbers of observation
points may slow down the performance.
4.13.3 File sizes

The size of the output files from D-WAQ PART simulations will depend on the length of the
simulation, the number of observation points and the number of output time intervals defined
for the zoom grid. The tutorial example <fti_tracer.inp> supplied with D-WAQ PART produces
about 13 Mbytes of output files.
54 of 137 Deltares
5 Tutorial
5.1 Introduction
This tutorial contains two cases which can be run with the D-WAQ PART module. The purpose
of the tutorial is to:
⋄ Give you experience with setting up a D-WAQ PART simulation, run the simulation and let
you understand what D-WAQ PART can simulate.
⋄ Check the installation of the D-WAQ PART-software.
T
This chapter gives both descriptive information about the steps to take to set up a D-WAQ PART
simulation and GUI-operations to complete each step. The GUI-operations are displayed in
the text boxes.
AF
5.2
5.2.1
D-WAQ PART tutorial cases
Tutorial cases
The tutorial contains two applications of 3D simulations with the Delft3D-FLOW ‘Friesian Tidal
Inlet’ schematisation. This schematisation contains 5 layers (aggregated from 10 hydrody-
namic layers) and 330 segments per layer.
The applications are (see also Table 5.1):

⋄ A simulation with two conservative tracers. One substance is released at the surface (layer
DR
1) as an instantaneous (or ‘dye’) release. The other substance is released continuously in

the bottom layer of the model (layer 5).
⋄ A simulation for two oil fractions (Ekofisk and Heavy Fuel Oil) that are released instan-
taneously and as floating substances (a phase of floating oil on the water surface). The
radius of the release is calculated with the oil-spill model. Dispersion of the oil is also
calculated with the oil-spill model.
Table 5.1: Specifications of the tutorial cases
Tutorial Case Tracer Tutorial Case Oil
Area ‘Friesian Tidal Inlet’ ‘Friesian Tidal Inlet’
Hydrodynamic database <com-tut_fti_part.dat> <com-tut_fti_part.dat>

<com-tut_fti_part.def> <com-tut_fti_part.def>
Type of substance Tracer(s) Oil
Substances ‘Tracer’ Ekofisk Oil

‘Tracer con 52’ Heavy Fuel Oil
D-WAQ PART input file <fti_tracer.mdp> <fti_oil.mdp>
Deltares 55 of 137
5.2.2 Starting D-WAQ PART

The D-WAQ PART model framework is started from the Delft3D-MENU as described in chap-
ter 3 (Getting started).
1. STARTING THE D ELFT 3D-MENU
Create a new directory first

⋄ Create a new directory somewhere on your hard-disk where you want to store the
T
results of the Particle tracking simulation.
⋄ Copy the <com-tut_fti_part.dat> and <com-tut_fti_part.def> files from your
<. . . /tutorial/part/friesian_tidal_inlet/all_flow_output> directory to this freshly created
directory.
AF ⋄
⋄
⋄
Start Delft3D
Start Delft3D-MENU.
Click Part.
Click the Select working directory button, the Select working directory window appears.
⋄ Browse to your freshly created directory and enter this directory.
⋄ Click OK to close the selection of your current working directory.
5.3 Coupling the hydrodynamic results

DR
5.3.1 Coupling module

For both tutorials, the same hydrodynamic database is used. The database needs to be
converted using the coupling program to make the necessary <∗.hyd>-file. The <∗.hyd>-
file can then be used for both cases.
⋄ Select the coupling program of D-WAQ PART by clicking Coupling and then Define input.
You are now accessing the GUI of the coupling module that will be described in the following
section.
2. STARTING THE COUPLING GUI
In the Particle tracking-selection window, click Coupling and next Define input.
5.3.2 Definition of the input

The three data groups which are enabled are:
⋄ Description
⋄ Hydrodynamics
⋄ Dispersion
The menu-items File and Help are also available.
56 of 137 Deltares
Tutorial
Description
With Description you can add a file description in order to produce some meta-information
about the coupling. Three lines of descriptive information are available, see Figure 5.1.
3. EDITING THE DESCRIPTION DATA GROUP
In the Delft3D-COUPLE window, click Description.

Enter 3 lines (of max. 40 characters) of meta-information.
T
AF
Figure 5.1: Coupling GUI - Data Group Description
DR
Hydrodynamics
With the Data Group Hydrodynamics, the hydrodynamic database is selected. By select-
ing Com-file, you can browse to and select the <com-tut_fti_part.dat> file, which can be
found in the <. . . /tutorial/part/friesian_tidal_inlet/all_flow_output> directory or in your cur-
rent working directory (it was copied here, see section 5.2.2). By clicking Open, the <com-
tut_fti_part.dat> file will be loaded into the GUI and information about this file will be dis-
played.
Aggregation in time
The timers in the file will automatically appear in the Time information frame. Time steps
can be increased, for example in order to reduce the required disk space for storage. Here to,
adapt the time step in the menu, but increase it only with an integer number (2 times, 3 times
etc.).
In this tutorial we do not aggregate in time, so no adjustment of timers is needed.
Horizontal aggregation
The grid can be aggregated in the horizontal within the frame Horizontal aggregation. The
three different options are:
⋄ No aggregation : no spatial aggregation is used
⋄ Use aggregation file (only Delft3D-WAQ);
⋄ Remove inactive cells: all dry segments are deleted automatically.
Here we choose option: Remove inactive cells
Deltares 57 of 137
Vertical aggregation
Layers can be aggregated in the vertical using the layer editor in the frame Vertical aggre-
gation. This tutorial case contains 10 (hydrodynamic) layers. If no vertical aggregation is
chosen, the number of water quality layers is equal to the number of hydrodynamic layers. To
reduce storage space and decrease calculation time, the number of water quality layers can
be reduced. To do so, select Edit layers. Figure 5.2a will appear.
T
AF
(a) Not aggregated (b) Aggregated
Figure 5.2: Vertical aggregation using the Layer editor. Left not aggregated and right
DR
aggregated layers.
On the left part of the Layer editor window, the number of water quality layers is shown. On
the right-hand side, the corresponding number of hydrodynamic layers is shown. By changing
the number in the input boxes (more than one hydrodynamic layer are combined to one water
quality layer), the number of water quality layers decreases.
Create 5 water quality layers by (starting from the top) changing the number of hydrodynamic
layers according to Figure 5.2b.
Remarks:
⋄ The aggregation shown in Figure 5.2b defines the link between the hydrodynamic and
water quality layers as follows:
water quality layer 1 = 2 hydrodynamic layers (1 and 2)
This is referred to as a ‘2-2-2-2-2’ aggregation.
⋄ The number of hydrodynamic layers remains intact (2+2+2+2+2=10).
The default situation when converting the hydrodynamic results is that there is NO vertical
aggregation.
58 of 137 Deltares
Tutorial
4. SPECIFYING AND AGGREGATING THE HYDRODYNAMIC DATABASE
In the Delft3D-COUPLE GUI, click Hydrodynamics.

Click Com-file, browse to the <. . . /tutorial/part/friesian_tidal_inlet/all_flow_input> direc-
tory or your current working directory and select the <com-tut_fti_part.dat> file.
Click Open to confirm (note that the details will be displayed in the Time information frame).
In the Horizontal aggregation frame, select Remove inactive cells (e.g. D-WAQ PART) from
the Type of: drop-down box.
T
In the Vertical aggregation frame, click Edit layers.
From top to bottom, change in the edit fields the value into “2”.
Click OK to confirm.
AF
As result your screen looks like Figure 5.3
DR
Figure 5.3: Coupling GUI - Data Group Hydrodynamics
Dispersion
Only the water quality model D-Water Quality may use the vertical diffusion from the hydro-
dynamics. D-WAQ PART ignores them because the values point to exchanges (= transition
from one cell to another). The random walk model used in D-WAQ PART is not dealing with
these exchanges, so dispersion values are ignored. Default values (for D-Water Quality) are
displayed. For the tutorial cases, no vertical diffusion coefficients are selected, see Figure 5.4.
5. EDITING DISPERSION
In the COUP-GUI, click Dispersion.

Do not check the Use minimum values check box.
Deltares 59 of 137
T
Figure 5.4: Coupling GUI - Data Group Dispersion
5.3.3 Saving input and running the coupling module

AF By selecting File → Save (as) you are requested to save the file. Your current working di-
rectory is the correct location. Enter the filename <com-tut_fti_part.hyd>. The extension
<hyd> will be given automatically. Selecting File → Exit to leave the COUP-GUI; you will
be back in the Hydrodynamic coupling selection window. Click Start to start the coupling.
A window will open in which the coupling of the hydrodynamic database is carried out. The
Coupling program writes information about the coupling to this window. When the coupling
is done, you can read this information (Figure 5.5) and close the window. Result files of the
coupling will be available in your current working directory. An interesting file to inspect is
<couplnef.out>.
DR
Figure 5.5: Coupling report output window
60 of 137 Deltares
Tutorial
6. SAVING INPUT AND RUNNING THE COUPLING
In the COUP-GUI, select File → Save (as) and save the file as <com-tut_fti_part.hyd> in
your current working directory.
Exit the COUP-GUI by selecting File → Exit. Then select Start to start the coupling.
To inspect the results, click Return, then Reports and subsequently Coupling in the Delft3D-
MENU, the file which will be displayed is <couplnef.out>.
T
5.4 Case 1: Tracer
Now the hydrodynamics have been coupled, the input for a PART simulation can be created.
Step by step it will be explained how to set up the input for a PART simulation. Each data
group will be explained. In this way the input file (<fti_tracer.mdp>) will be created.
AF Select the Define input button in the selection window for Particle tracking (see Figure 5.6).
DR
Figure 5.6: Selection window for Particle tracking
This will start the PART-GUI (see Figure 5.7). With the GUI, you can input or edit data which
are necessary for preparing a D-WAQ PART scenario (input file). The related data are divided
in several data groups. From top to bottom, the following data groups are available: De-
scription, Hydrodynamics, Substances, Time Frame, Initial Condition, Numerical Parameters,
Instantaneous Releases, Continuous Releases, Process Parameters, Observation Points and
Output Options.
Deltares 61 of 137
T
AF
Figure 5.7: Main window of the PART-GUI
5.4.1 Description
DR
This data group allows you to add 3 lines of description (meta-information) about the D-
WAQ PART scenario. Specify the 3 lines as in Figure 5.8.
7. DATA GROUP : description
In the PART-GUI, select Description.

Specify the lines of meta data like Figure 5.8.
When finished, you can click on the next data group, Hydrodynamics.
62 of 137 Deltares
Tutorial
5.4.2 Hydrodynamics
Available hydrodynamic results (e.g. <∗.hyd>-files from the coupling module) can be se-
lected in the Data Group Hydrodynamics. Press Select File to open the Select hydrodynam-
ics file window. All the results found in the current working directory are listed; other results
can be found by browsing to a different directory. Select the file <com-tut_fti_part.hyd>. The
contents is listed in the Selected hydrodynamic results frame, Figure 5.9.
T
AF
Figure 5.9: Data Group Hydrodynamics showing the selected hydrodynamic results
<com-tut_fti_part.hyd>
DR
8. DATA GROUP : hydrodynamics
In the PART-GUI, select Hydrodynamics.

Press Select File and open the <com-tut_fti_part.hyd> file.
5.4.3 Substances
After selection of the Data Group Substances the window depicted in Figure 5.10 appears.
Select the Tracer radio button and use Add to add a second substances. In the list “(New)”
appears. Change the name of the substances in:
⋄ “Tracer con 52”
Deltares 63 of 137
T
AF Figure 5.10: Data group Substances
9. DATA GROUP : substances
In the PART-GUI, select Substances.

Select radio button Tracer in the Type of model frame.
Change the name in the Name of substance edit field into “Tracer”.
DR
Click Add.
Change the name in the Name of substance edit field into “Tracer con 52”.
Click in the list box.
5.4.4 Time frame

Figure 5.11 shows the window corresponding to the Time Frame Data Group. The time frame
of the hydrodynamic results is given. You can specify the following items:
⋄ Stop time
⋄ Time step
The start time for the water quality calculation cannot be changed. This is equal to the start
time of the coupled hydrodynamics.
Set the timings according to Table 5.2.
Table 5.2: Timers for the Tracer tutorial case
Tracer tutorial case
Start time 05/08/1990 12:30:00

Stop time 05/08/1990 20:30:00
Time step for calculation 15 minutes
64 of 137 Deltares
Tutorial
T
AF Figure 5.11: Data Group Time Frame.
10. DATA GROUP : time frame
In the PART-GUI, select Time Frame.

Specify the Stop time edit field: “05 08 1990 20 30 00”.
Specify the Time step edit field: “00 00 0000 00 15 00”.
DR
5.4.5 Numerical parameters

Figure 5.12 corresponds to the Data Group Numerical Parameters. The following numerical
parameters can be specified here:
⋄ the number of particles
⋄ vertical diffusion option
⋄ vertical diffusion scaling factor
Figure 5.12: Data Group Numerical Parameters after editting
In this tutorial case, the number of particles is 100 000.
Deltares 65 of 137
Two vertical diffusion options are available:

⋄ Depth averaged algebraic: the vertical diffusion coefficient is not a function of z , and its
integral is approximated by an analytical function. However, it is a function of the (x,y )
co-ordinates depending on the local wind stress and bed stress.
⋄ Constant dispersion coefficient: the vertical diffusion coefficient is a constant value for the
whole model domain.
The other two options (Algebraic and Import from hydrodynamics) are expected in the near
future, and will have the characteristics:
T
⋄ Algebraic: coefficient is a function of z and its integral will be approximated.
⋄ Import from hydrodynamics: coefficient is a function of z .
The vertical diffusion scaling factor can be set by you. Its default value is 1, meaning the
AF whole vertical diffusion coefficient is taken in the calculations. For the Tracer tutorial case, the
vertical diffusion scaling factor is 0.05.
11. DATA GROUP : numerical parameters
In the PART-GUI, select Numerical Parameters.

Specify “100 000” in the Number of particles edit field.
Specify the “Depth averaged algebraic” radio button in the Vertical dispersion frame.
Set the Scale factor in the Vertical dispersion frame to: “0.05”.
DR
5.4.6 Instantaneous releases

An instantaneous discharge of a substance is referred to as a ‘dye release’. In this data
group you can prepare (see Figure 5.13): a list of dye releases with corresponding locations
(co-ordinates), release radius, particles used (as % of the total number of particles), and
a layer number where particles are released (which is 1 for 2D, and can be chosen for 3D
calculations).
The location of an instantaneous release can be defined by using point-and-click in the Visu-
alisation Area, or the (x,y ,z ) co-ordinates of the location can be typed in on the main window
of this data group. Note that for these locations the particle model does not work with grid-cell
indices since these are not relevant hereto. The release time and mass of substance released
are given in a sub-window. Enter the values in Table 5.3, see Figure 5.13 and Figure 5.14.
66 of 137 Deltares
Tutorial
Table 5.3: Details of dye release 1
Description Value
release name dye release 01

location x [m] 193 049
location y [m] 609 122
layer 1
particles (% of total) 50
T
release radius [m] 0
release time 05/08/1990 12:30:00
substance released Tracer
amount released [kg] 1.0 · 108
AF
DR
Figure 5.13: Data Group Instantaneous Releases
Remarks:
⋄ In a 3D model the release is discharged uniform over the selected layer. The z co-
ordinate has no meaning.
⋄ In a 2D model the z co-ordinate is a relative co-ordinate and refers to the relative depth
below the water surface (0 % = surface, 100 % = bottom).
Deltares 67 of 137
T
Figure 5.14: Release tables for instantaneous releases window of a selected instanta-
neous release
AF
To use the point-and-click method, do not use Add, but select View → Visualisation Area from
the menu bar. Set the Edit Mode to Add, see Figure 5.15. If not set yet, select Instantaneous
releases in the Edit menu. To define a release at a specific location, move around with the
cursor till you reach the specific location and then click with the left mouse. In the main window
of the User Interface change the default name ‘(new)’ to ‘dye release 01’ and set the other
characteristics according to Table 5.3.
DR
Figure 5.15: Visualisation Area window
12. DATA GROUP : instantaneous releases
In the PART-GUI, select Instantaneous Releases.

Click Add.
Specify “dye release 01” in the Instantaneous release edit field.
Specify “193049” in the X edit field.
Specify “609122” in the Y edit field.
Select layer “1” from the Layer drop-down box.
Specify “50” in the Particles used edit field.
68 of 137 Deltares
Tutorial
Specify “0” in the Radius edit field.

Select dye release 1 (from the list) and click Release Tables.
In the Release tables for instantaneous releases window, specify (see Figure 5.14).
⋄ “05/08/1990 12:30:00” in the Release time edit field (Time frame box)
⋄ “1.000e+008” in the Tracer edit field (Substances frame).
⋄ Click Close to return to the main window.
T
5.4.7 Continuous releases
In this data group (Figure 5.16 shows the corresponding window) you can prepare a list of
AF continuous releases with corresponding locations, release radius, particles used (as % of the
total particles), and a layer number (which is 1 for 2D, and can be chosen for 3D models).
The co-ordinates of the continuous release can be obtained with the Visualisation Area or
can be typed in on the main window of this data group in (x, y)-co-ordinates. Note that for
these locations the particle model does not work with grid-cell indices since these are not
relevant for particle models. The release rates, stoichiometric coefficients per substance, and
scale factor are given in a sub-window.
Table 5.5: Details of the continuous release

DR
Description Value
release name cont. release 01

location x [m] 193567
location y [m] 605871
layer 5
particles (% of total) 50
release radius [m] 0
scale factor 1
3
release rate [m /s] 1.0
substance Tracer con 52
3
concentration [kg/m ] 1.0 · 105
Deltares 69 of 137
T
AF Figure 5.16: Data Group Continuous Releases
DR
Figure 5.17: Release tables for continuous releases window of continuous release 1
13. DATA GROUP : continuous releases
In the PART-GUI, select Continuous releases.

Click Add.
Specify “continuous release 1” in the Continuous release edit field.
Specify “193 567” in the X edit field.
Specify “605 871” in the Y edit field.
Select layer “5” from the Layer drop-down box.
Specify “0” in the Radius edit field.
Select continuous release 1 (from the list) and click Release Tables.
70 of 137 Deltares
Tutorial
13. DATA GROUP : continuous releases
In the Release tables for continuous releases window, specify (see Figure 5.17):
⋄ Specify in the first time breakpoint row: “05 08 1990 12 30 00” in the first column and
“1.000” in the second column [m3 /s].
⋄ Specify in the second time breakpoint row: “05 08 1990 20 30 00” in the first column and
“1.000” in the second column [m3 /s].
⋄ Select interpolation method “linear” (radio button).
⋄ Specify Scale factor : “1.0”.
T
⋄ “1.000e+005” in the Tracer con 52 edit field (Concentration frame).
AF
5.4.8 Process parameters
In the Data Group Process Parameters you can define:
⋄ first order decay rates [d−1 ] for mass decay of substances
⋄ parameters for sedimentation and erosion of particles
⋄ physical parameters (for wind, bed stress and diffusion)
⋄ oil specific parameters (oil model only)
Decay rates
Decay rates are given in a table, per substance and per time breakpoint in units per day. Click
DR
Decay Rates to display the table (see Figure 5.18).
Figure 5.18: Data Group Process Parameters with sub-data group Decay rates
Deltares 71 of 137
Sedimentation/Erosion
When clicking Sedimentation/Erosion, two additional sub-data groups appear, namely:
⋄ Constants
⋄ Settling velocities
First, you must specify if sedimentation and erosion from the bottom is included or excluded
(see Figure 5.19). When erosion is included, 3 parameters must be specified for sedimenta-
tion and erosion according to Partheniades (1965) and Krone (1962). In the tutorial examples
T
this is not further discussed, since erosion is excluded.
AF
DR
Figure 5.19: Sub-data group Sedimentation/erosion – Constants
Settling velocities are given with 6 parameters (two amplitudes, (a0 and a1 ); period (ω ), phase
(φ), minimum (vmin ) and maximum (vmax ) settling velocities) and are also given in tables.
Each of the 6 parameters has its own table, per substance and time breakpoint.
72 of 137 Deltares
Tutorial
T
AF
Figure 5.20: Settling Velocities details (A0 list box selected)
For this tutorial case, these parameters must be specified as given in Table 5.7.
Table 5.7: Process parameters

DR
Description Parameter Value
Decay No decay
Settling Velocities A0 [m/s] 1.0 · 10−4 (Tacer)

0.0 (Tracer con 52)
A1 [m/s] 0.0
period [h] 0.0
phase [h] 0.0
Vmin [m/s] -10
Vmax [m/s] 100
Physical Parameters coefficient a 0.07

coefficient b 0.7
roughness [m] 0.002
wind drag [%] 3.0
3
density of water [kg/m ] 1024.0
wind speed [m/s] 1
wind direction [degrees] 270
Deltares 73 of 137
Physical
So-called physical process parameters contain constants as well as time dependent parame-
ters of wind and horizontal dispersion. See Figure 5.21 and Table 5.7.
T
AF
Figure 5.21: Physical Process parameters
DR
14. DATA GROUP : process parameters
In the PART-GUI, select Process Parameters.

Click Decay rates.
Specify the table according to Figure 5.18.
Click Sedimentation/erosion.
Click Constants.
Select the Exclude sedimentation and erosion processes radio button.
Click Settling Velocities.
Specify each item according Table 5.7.
Click Physical.
Specify each item according Table 5.7.
5.4.9 Observation points

To have a time history of results at a single point, observation points can be selected (the
corresponding data group window is presented in Figure 5.22). A list of observation points,
i.e. their names and their co-ordinates can be given here.
In this tutorial case, a single observation point has been chosen (‘waste point’) at location:
x = 193 567; y = 605 871
Observation points can also be added with the Visualisation Area.
74 of 137 Deltares
Tutorial
T
AF Figure 5.22: Data Group Observation Points
15. DATA GROUP : observation points
In the PART-GUI, select Observation points.

Click Add.
Change the name of the observation point into: “waste point”.
DR
5.4.10 Output options

This data group (see Figure 5.23) contains the time frames for standard output results (history
file and map file) and the following additional output options:
⋄ A zoom grid, together with its rectilinear grid dimensions.
⋄ The Zoom Grid Output.
The Zoom Grid Output must be set to specific moments in time using the table displaying the
time breakpoints.
(Zoom grid) Grid definition

The zoom grid should be defined in the following way (4 edit fields):
⋄ X: from 185 000 (lower) to 210 000 (upper)
⋄ Y: from 600 000 (lower) to 625 000 (upper)
⋄ Number of cells in both x- and y -direction: 150.
Deltares 75 of 137
T
AF
Figure 5.23: Data Group Output Options, with also showing options for Grid Definition of
the Zoom grid
16. DATA GROUP : output options
In the PART-GUI, select Output Options.

DR
Change (if necessary) the content of the Time frame according Figure 5.23.
Click Grid Definition in the Zoom grid frame.
Specify “185000” in the X lower edit field.
Specify “210000” in the X upper edit field.
Specify “600000” in the Y lower edit field.
Specify “625000” in the Y upper edit field.
Specify “150” in the Number of cells in X-direction edit field.
Specify “150” in the Number of cells in Y-direction edit field.
Click Output Definition.
Specify (using menu-item Table → Insert row) the number of breakpoints table according
Table 5.8.
76 of 137 Deltares
Tutorial
Table 5.8: Time breakpoints definition
time breakpoints recovery factor

dd mm yyyy hh mm ss
05 08 1990 12 30 00 1.000
05 08 1990 12 45 00 1.000
05 08 1990 13 00 00 1.000
.. ..
T
. .
05 08 1990 20 15 00 1.000
05 08 1990 20 30 00 1.000
AF (Zoom grid) Output definition
The tutorial case has 8 time breakpoints for which output is written to this zoom grid. These
time breakpoints should be defined (see Figure 5.24). Select a row in the time breakpoints ta-
ble and select the menu-item Table → Insert row. Now a row is added to the time breakpoints
table. Change the content of the newly added row according to Table 5.8, for this tutorial we
use the button Generate table.
DR
Figure 5.24: Zoom Grid Output
5.4.11 Saving the input

By selecting the menu-item File → Save as. . . you can specify a file name for your PART
simulation. Suggestion is to name it <fti_tracer.mdp>. It will be automatically stored in
your current working directory. Select menu-item File → Exit, ignore the message when you
saved your work already. Now your input file is ready and the simulation can take place (see
section 5.6).
Deltares 77 of 137
5.5 Case 2: Oil

After the hydrodynamics have been coupled (see section 5.3), the input for a PART-simulation
can be created. The hydrodynamic database is the same as in the Tracer tutorial case. Step
by step it will be explained how to set-up the input for a PART-simulation (oil module). Each
data group will be explained. In this way the input file (<fti_oil.mdp>) will be created.
Select Define input in the selection window for Particle tracking, see Figure 5.25.
T
AF
DR
Figure 5.25: Selection window for Particle tracking
This will start the PART-GUI (see Figure 5.26). With the GUI you can input or edit data which
are necessary for preparing a D-WAQ PART scenario (input file). The related data are divided
in several data groups. From top to bottom, the following data groups are available: Descrip-
tion, Hydrodynamics, Substances, Time Frame, Initial Conditions, Numerical Parameters, In-
stantaneous Releases, Continuous Releases, Process Parameters, Observation Points and
Output Options.
78 of 137 Deltares
Tutorial
T
AF
Figure 5.26: Main window of the PART-GUI
5.5.1 Description
DR
This data group allows you to add 3 lines of description (meta-information) in a D-WAQ PART
input file. Specify the 3 lines as in Figure 5.8.
1. DATA GROUP : description
In the PART-GUI, select Description.

Specify the lines of metadata like Figure 5.27.
When finished, you can click on the next data group, Hydrodynamics.
Deltares 79 of 137
5.5.2 Hydrodynamics
Available hydrodynamic results (e.g. <∗.hyd>-files from the coupling module) can be se-
lected in the Data Group Hydrodynamics. All the results found in the current working directory
are listed, whereas other results can be found by browsing to a different directory. The file
<com-tut_fti_part.hyd> is already listed in the Select hydrodynamics file window.
T
AF
Figure 5.28: Data Group Hydrodynamics showing the selected hydrodynamic results
<com-tut_fti_part.hyd>
DR
2. DATA GROUP : hydrodynamics
In the PART-GUI, select Hydrodynamics

The <com-tut_fti_part.hyd> file is listed in the Select hydrodynamics file window.
5.5.3 Substances
After selection of the Substances Data Group the window depicted in Figure 5.29 appears.
Select the Oil radio button, the oil type “Ekofisk” is immediately available, use the Add button
to add a second substance. In the list (new) appears. Change the name of the substance in:
⋄ “Heavy Fuel Oil”
⋄ Click in the list box
Note that oil substances are always given in three fractions, namely:
⋄ Floating oil: oil that is floating as a thin layer on top of the water column,
⋄ Dispersed oil: oil that is in the water column (‘dissolved’ oil droplets),
⋄ Sticking oil: oil that (may) stick to land or to the bed, this fraction is no longer transported
by water.
80 of 137 Deltares
Tutorial
T
AF Figure 5.29: Data Group Substances
3. DATA GROUP : substances
In the PART-GUI, select Substances.

Select radio button Oil in the Type of model frame.
Change name in the Name of substance edit field into “Ekofisk”.
DR
Click Add.
Change name in the Name of substance edit field into “Heavy Fuel Oil”.
Click in the list box.
5.5.4 Time frame

Figure 5.30 shows the window corresponding to the Time Frame Data Group. The time frame
of the hydrodynamic results is given. You can specify the following items:
⋄ Stop time
⋄ Time step
The start time for the water quality calculation cannot be changed. This is equal to the start
time of hydrodynamic calculation (after coupling).
Deltares 81 of 137
Table 5.9: Timers for the Oil tutorial case
Oil tutorial case
Start time 05/08/1990 12:30:00

Stop time 05/08/1990 20:30:00
Time step for calculation 15 minutes
T
AF
DR
Figure 5.30: Data Group Time Frame
4. DATA GROUP : time frame
In the PART-GUI, select Time Frame.

Specify the Stop time edit field: “05 08 1990 20 30 00”.
Specify the Time step edit field: “00 00 0000 00 15 00”.
5.5.5 Initial condition

In this Data Group you select whether or not you want to specify the extent of initial oil patches,
by selecting an initial patch file. The file specifies the extent, the amount of oil for each patch
and the number of particles to represent each patch of oil. The number of particles specified
here are in addition to those specified in the Numerical Parameters Data Group. The format
of the initial condition file is specified in section A.6.
82 of 137 Deltares
Tutorial
Figure 5.31: Data Group Initial Condition
T
Press Select file to open the file with initial patches, see Figure 5.32.
AF
DR
Figure 5.32: Selection window for the initial conditions file
5. DATA GROUP : Initial Condition
In the PART-GUI, select Initial Conditions.

Select radio button Initial oil patches.
Press the Select file button.
Select file <ekofisk_hfo.ini>.
Click Open.
Deltares 83 of 137
5.5.6 Numerical parameters

Figure 5.33 corresponds to the Numerical Parameters Data Group. The following numerical
parameters can be set here:
⋄ number of particles
⋄ vertical dispersion option
⋄ vertical dispersion scaling factor
T
AF
Figure 5.33: Data Group Numerical Parameters
DR
In this tutorial case, the number of particles is 200 000.
Two vertical dispersion options are available:

⋄ Depth averaged algebraic: the vertical dispersion coefficient is not a function of z , and its
integral is approximated by an analytical function. However, it is a function of the (x, y)
co-ordinates depending on the local wind stress and bed stress.
⋄ Constant dispersion coefficient: the vertical dispersion coefficient is a constant for the
whole model domain.
The other two options (Algebraic and Import from hydrodynamics) are expected in the future,
and will have the characteristics:
⋄ Algebraic: coefficient is a function of z and its integral will be approximated.
⋄ Import from hydrodynamics: coefficient is a function of z .
The vertical dispersion scaling factor can be set by you. Its default value is 1, meaning the
whole vertical dispersion coefficient is taken in the calculations. For the Oil tutorial case, the
vertical dispersion scaling factor is 0.1.
6. DATA GROUP : numerical parameters
In the PART-GUI, select Numerical Parameters.

Specify “200000” in the Number of particles edit field.
Specify the “Depth Averaged algebraic” radio button in the Vertical dispersion frame.
Set the Scale factor in the Vertical dispersion frame to: “0.1”.
84 of 137 Deltares
Tutorial
Table 5.10: Details of oil releases
Description Value
release name oil patch 1 oil patch 2

location x [m] 193049 193567
location y [m] 609122 605871
particles (% of total) 50 50
T
release radius [m] calculated calculated
release time 05/08/1990 12:30:00 05/08/1990 12:30:00
substance released Ekofisk Heavy Fuel Oil
6
amount released 1.0 · 10 [kg] 1.0 · 106 [kg]
AF
5.5.7 Instantaneous releases
In this data group you can prepare (see Figure 5.34): a list of instantaneous releases with cor-
responding locations (co-ordinates), release radius, particles used (as % of the total number
of particles), and a layer number where particles are released (which is 1 for 2D, and can be
chosen for 3D calculations).
The co-ordinates of the instantaneous release can be obtained with the Visualisation Area
by point-and-click. You can also enter in the main window of this data group the (x, y, z)
DR
co-ordinates. Note that for these locations, as with all location references in PART, the particle
model does not work with grid-cell indices since these are not relevant hereto. The release
time and mass of substance released are given in a sub-window:
Remark:
⋄ The volume can be calculated by dividing the total discharge (1.0·106 kg) by the density
(850 kg/m3 for oil patch 1 and 990 kg/m3 for oil patch 2, see section 5.5.9).
Figure 5.34: Data Group Instantaneous Releases
Deltares 85 of 137
Remark:
⋄ Oil can only be released at the water surface (Layer 1, top of that layer). The z -value
edit field and Layer drop-down box are disabled.
T
AF Figure 5.35: Release tables for instantaneous releases window for oil patch 1
DR
Figure 5.36: Release tables for instantaneous releases window for oil patch 2
In the PART-GUI, select Instantaneous Releases.

Click Add.
Specify “oil patch 1” in the Instantaneous release edit field.
Select radio button Calculated radius.
Click Add (again).
Specify “oil patch 2” in the Instantaneous release edit field.
Select radio button Calculated radius.
86 of 137 Deltares
Tutorial
Select oil patch 1 (from the list) and click Release Tables.
In the Release tables for instantaneous releases window, specify (see Figure 5.35).
⋄ “05/08/1990 12:30:00” in the Release time edit field (Time frame box).
⋄ “1.000e+006” in the Ekofisk edit field (Substances box).
Select oil patch 2 (from the list) and click Release Tables.
T
In the Release tables for instantaneous releases window, specify (see Figure 5.36):
⋄ “05/08/1990 12:30:00” in the Release time edit field (Time frame box).
⋄ “1.000e+006” in the Heavy Fuel Oil edit field (Substances box).
AF
5.5.8 Continuous releases
There are no continuous releases in this tutorial case.
5.5.9 Process parameters

In the Data Group Process Parameters (see Figure 5.37) you can define:
⋄ first order decay rates [d−1 ] for mass decay of substances
DR
⋄ parameters for sedimentation and erosion of particles

⋄ physical parameters (for wind, bed stress and diffusion)
⋄ oil specific parameters (oil model only)
Decay rates
Decay rates are given in a table, per substance and per time breakpoint in units per day. For
oil you can specify decay rates per fraction (floating, dispersed, sticked). Click Decay Rates
to inspect the table. All values should be zero (use the horizontal scroll bar).
Deltares 87 of 137
T
AF
Figure 5.37: Data Group Process parameters with sub-data group Decay rates
Sedimentation/Erosion
When clicking Sedimentation/Erosion, two additional sub-data groups appear, namely:
⋄ Constants
⋄ Settling Velocities
DR
First, you must specify if sedimentation or erosion from the bottom is included or excluded
(see Figure 5.38). When erosion is included, parameters must be specified for sedimentation
and erosion according to Partheniades (1965) and Krone (1962). In the tutorial examples this
is not further discussed, since erosion is excluded.
Figure 5.38: Sub-data group Sedimentation/Erosion
Settling velocities are given with 6 parameters (two amplitudes, (a0 and a1 ); period (ω ), phase
(φ), minimum (vmin ) and maximum (vmax ) settling velocities), and are also given in tables.
Each of the 6 parameters has its own table, per substance and time breakpoint.
88 of 137 Deltares
Tutorial
Table 5.12: Process parameters for Sedimentation and Erosion
Settling Velocities for A0 [m/s] −1.0 · 10−4

Ekofisk dispersed and A1 [m/s] 0.0
HFO dispersed period [h] 0.0
phase [h] 0.0
T
Vmin [m/s] -10
Vmax [m/s] 100
Remark:
AF
⋄ The time breakpoints most be the same for all parameters.
DR
Figure 5.39: Settling Velocities details (A0 selected in list box)
For this tutorial case, these parameters must be specified as given in Table 5.14. The negative
settling velocity for dispersed oil gives the oil buoyancy.
Deltares 89 of 137
Table 5.13: Physical process parameters
Physical Parameters coefficient a 0.07

coefficient b 0.7
roughness [m] 0.002
wind drag (%) 3.0
3
T
density of water [kg/m ] 1024
wind speed [m/s] 6
AF wind direction [degrees] 270
Physical
So-called physical process parameters contain constants as well as time dependent param-
eters of wind and horizontal diffusion, see Figure 5.40. These parameters must be specified
as given in Table 5.13.
DR
Figure 5.40: Physical Process parameters
90 of 137 Deltares
Tutorial
Table 5.14: Process parameters for oil
Ekofisk HFO
Oil Model Evaporation per day 0.4 0.05

Stickyness probability 0.0 0.5
Volatile fraction 1 0.94
T
Emulsification parameter C1 0 2 · 10−6
Maximum water content 1 0.7
Fraction at which emulsification starts 1 0.05
Density 850 990
AF Kinematic viscosity
Dispersion (entrainment)
50 1500
calculated calculated
Minimum thickness 0.00005 0.00005
Deflection angle 0 0
Oil Model
In this part, specific oil parameters are specified concerning dispersion, evaporation, emulsi-
DR
fication, etc., see Figure 5.41. Fill in the values as indicated in Table 5.14.
Figure 5.41: Oil Model specific parameters
Deltares 91 of 137
8. DATA GROUP : process parameters
In the PART-GUI, select Process Parameters.

Click Sedimentation/erosion.
Click Constants.
Select the Exclude sedimentation and erosion processes radio button.
Click Settling Velocities.
Specify each item according to Table 5.12.
T
Click Physical.
Specify each item according to Table 5.13/Figure 5.40.
Click Oil Model.
Specify each item according to Table 5.14/Figure 5.41.
AF
5.5.10 Observation points
To have a time history of results at a single point, observation points can be selected (the
corresponding data group window is presented in Figure 5.22. A list of observation points, i.e.
their names and their co-ordinates can be given here.
DR
Figure 5.42: Data Group Observation Points
In this tutorial case, a single observation point has been chosen (‘waste point oil’) at location:
x = 193 567; y= 605 871
Observation points can also be added with the Visualisation Area.
92 of 137 Deltares
Tutorial
9. DATA GROUP : observation points
In the PART-GUI, select Observation Points.

Click Add.
Change the name of the observation point into: “waste point oil”.
T
5.5.11 Output options
This data group (see Figure 5.43) contains the time frames for standard output results (history
file and map file) and additional output mapped on a rectangular zoom grid.
AF The output to the zoom grid must be set to specific moments in time using the table displaying
the time breakpoints.
DR
Figure 5.43: Data Group Output Options, also showing options for Definition of Zoom grid
(Zoom grid) Grid definition

The zoom grid should be defined in the following way (4 edit fields):
⋄ X: from 185000 (lower) to 210000 (upper)
⋄ Y: from 600000 (lower) to 625000 (upper)
⋄ Number of cells in x- and y -direction is: 150.
Deltares 93 of 137
(Zoom grid) Output definition

The tutorial case has 8 time breakpoints for which output is written to this zoom grid. These
time breakpoints should be defined (see Figure 5.44). Select a row in the time breakpoints ta-
ble and select the menu-item Table → Insert row. Now a row is added to the time breakpoints
table. Change the content of the newly added row according to Table 5.15, for this tutorial we
use the button Generate table.
T
AF
DR
Figure 5.44: Zoom Grid Output
Table 5.15: Time breakpoints definition
time breakpoints recovery factor

dd mm yyyy hh mm ss
05 08 1990 12 30 00 1.000
05 08 1990 12 45 00 1.000
05 08 1990 13 00 00 1.000
.. ..
. .
05 08 1990 20 15 00 1.000
05 08 1990 20 30 00 1.000
5.5.12 Saving the input

By selecting the menu-item File → Save as. . . you can specify a file name for your PART
simulation. Suggestion is to name it <fti_oil.mdp>. It will be automatically stored in your
current working directory. Select menu-item File → Exit and ignore the message when you
already saved your work. Now your input file is ready and the simulation can take place (see
section 5.6).
94 of 137 Deltares
Tutorial
5.6 Executing a scenario

Once you leave the PART-GUI, you will return to the main menu. The calculation can be
started by selecting Start from the main menu screen (Figure 5.6).
The model will always run with the latest input file that has been opened and saved.
While the model is running, you will see on the screen the hydrodynamic flow and volume
records updated, the actual time, and the number of active particles in the calculation. When
the calculation is finished, you will again see the main menu of D-WAQ PART.
T
When the model stops before the end of simulation without an error message, then check the
report file on the word ‘ERROR’. Errors in the syntax of the input file must appear directly on
the screen in front of you. There should be no errors and no warnings for the two tutorial
cases.
5.7
AF Viewing the report
A report of each model calculation is made, and this can be viewed by selecting Reports
from the main menu screen. By clicking Monitoring and Select file, you can browse to the
<∗.out> file (∗ = name of scenario). You can scroll through the file to read any error or
warning messages for the D-WAQ PART calculation.
The possible error and warning messages are supposed to be instructive and clear enough for
you to make the necessary corrections. Look for the words ‘ERROR’ and ‘WARNING’ (capital
letters). There should be no errors and no warnings for the two tutorial cases.
DR
5.8 Viewing the results

To view output of a PART simulation, the Delft3D-QUICKPLOT or GPP (General Post Pro-
cessing) program must be used. By selecting QUICKPLOT or GPP in the selection window
for Particle tracking (Figure 5.6), QUICKPLOT or GPP starts.
GPP:
For GPP to open the so-called session file (<∗.ssn>), the datasets and plots of the tutorial
examples will be opened. The session files to open are <tutorial_part_fti_tracer.ssn> or
<tutorial_part_fti_oil.ssn>. Both can be found in the <. . . \tutorial\part\friesian_tidal_inlet>
directory. Use the menu-item Session → Open. . . (click “No” if you are asked to save
changes) to browse to the required file, click Open to confirm.
The plots can be found when you click Plots. Select a plot from the list and click View → Edit
to open it.
Remarks:
⋄ The ‘tracer’ plot can only be viewed if you have finished the tutorial Tracer case.
⋄ The ‘oil’ plot can only be viewed if you have finished the tutorial Oil case.
Deltares 95 of 137
T
AF
DR
Figure 5.45: Tracer Case: Instantaneous release: situation at 18:00 hrs (upper) and
19:30 hrs (lower) in the surface layer
96 of 137 Deltares
Tutorial
T
AF
DR
Figure 5.46: Tracer Case: Continuous release: situation at 20:30 hrs in layer 3: MAP-file
(upper) and PLO-file (lower)
Deltares 97 of 137
T
AF
DR
Figure 5.47: Oil Case: Situation at 19:30 hrs in the surface layer: Ekofisk dispersed (up-
per) and Floating HFO (lower)
98 of 137 Deltares
Tutorial
T
AF
DR
Figure 5.48: Oil Case: Situation at 17:00 hrs (upper) and 19:30 hrs (lower), PLO-file for
Floating HFO in the surface layer
Deltares 99 of 137
T
AF
DR
100 of 137 Deltares

6 Conceptual model
6.1 Introduction
D-WAQ PART is a 3-dimensional particle tracking model that is particularly useful for mid-
field water quality modelling. It calculates a dynamic concentration distribution by following
the tracks of thousands of particles with time. The model provides a detailed description of
concentration distributions, resulting from instantaneous or continuous releases of salt, oil,
temperature or other conservative or simple decaying substances. This chapter gives you a
brief introduction to the computer model and its applications (see also WL | Delft Hydraulics
T
(1993); Van den Boogaard et al. (1993)).
6.1.1 Operational functionality

D-WAQ PART can operate in the following modes:
AF ⋄ Standard mode: 3D or 2DH mode: D-WAQ PART coupled to 3D or 2DH Delft3D-FLOW
(one layer model or multi-layer model, sigma-layers only, Z -layers is not supported by D-
WAQ PART), and extended with an analytical vertical velocity profile for bottom shear and
wind.
⋄ Oil spill module: simulation of oil spills with floating and dispersed oil fractions.
6.1.2 Physical system

The D-WAQ PART model is used for water quality modelling. D-WAQ PART simulates trans-
port processes and simple chemical reactions of substances. The model can simulate de-
DR
tailed shapes of patches of wasted material.
The physical components in the system are:

⋄ The water system: a lake, estuary, harbour or river, possibly with open boundaries to other
water systems. Tidal variations are included.
⋄ Outfalls due to human activities.
⋄ Chemical substances like rhodamine dyes, salt, oil or a demand of oxygen due to fast
chemical reactions.
⋄ Wind fields.
⋄ Dissolved and/or particulate substances.
6.1.3 Processes
This section describes the D-WAQ PART model in terms of physical processes or phenomena
that the model was designed to represent.
⋄ The dynamics of patches close to an outfall location.
⋄ Simple first-order decay processes like the decay of several fractions of oil.
⋄ Vertical dispersion for well-mixed systems.
⋄ Horizontal dispersion due to turbulence. According to turbulence theory this dispersion
increases in time.
⋄ The effects of time-varying wind fields on the patches.
⋄ The effects of bottom-friction on the patches.
⋄ The existence of a plume at the outfall (rather than a point-source) by starting the simula-
tion from a circular plume with an estimated or field-measured radius.
⋄ Settling and resuspension of particles.
Deltares 101 of 137

6.2 Principles of a random walk particle tracking model

D-WAQ PART is a random walk particle tracking model, which is based on the principle that
the movement of dissolved (or particulate) substances in water can be described by a limited
(large) number of discrete particles that are subject to advection due to the currents and by
horizontal and vertical dispersion. The movement of the particles consists therefore of two
elements. For each time-step, the first step is the advection step due to the shear stresses
from currents (bottom) and wind (surface). The second step is the random walk step in which
the size and direction of the movement is a random process but is related to the horizontal
and vertical dispersion. More details are given in chapter 7.
T
6.3 Dispersion coefficients
6.3.1 Introduction
AF The dispersion coefficients are the input parameters for the random spreading processes. In
the horizontal and vertical direction, the dispersion coefficients are treated separately, even
through the numerical treatment is the same.
6.3.2 Horizontal dispersion

The horizontal dispersion coefficient accounts for all of the random fluctuations in the horizon-
tal that are not resolved in the Delft3D-FLOW hydrodynamics. For particle tracking simula-
tions, the horizontal dispersion coefficient is also time dependent. In the initial period of time
after the release of the particles, the patch of particles is relatively small and the mixing of
the particles is caused by small-scale turbulence effects only. However, after some time, the
DR
‘cloud’ of particles will have spread sufficiently such that larger-scale eddies and circulations
will now contribute to the mixing effect.
The dispersion coefficient models the effects of turbulence that were not included in the hy-
drodynamics.
Turbulence model
Turbulent diffusion accounts for small scale deviations from the mean hydrodynamic velocity
field. These deviations are partially correlated and therefore lead to a net extra displacement.
This net extra displacement has a random direction. Therefore it is appropriate to model the
effect as dispersion.
Since D-WAQ PART is a dynamic 3D model dispersion coefficients are small (order 1 m2 /s).
The turbulence processes on this scale are time-dependent. We will demonstrate this by
using kinetic gas-theory (Csanady, 1973), describing molecular diffusion E :
Z t
E= ⟨⃗v (0), ⃗v (τ )⟩ dτ (6.1)
0
The bracketed term under the integral is named the Lagrangian correlation coefficient. Using
the theory of Brownian motion it can be approximated as:
⟨⃗v (0), ⃗v (τ )⟩ = |v0 |2 e−τ /tL (6.2)
with tL an appropriate time-scale. This expression leads to a diffusion coefficient of the form:
E = v02 tl (1 − exp(−t/tl )) (6.3)
102 of 137 Deltares

Conceptual model
Note that the two limits of this function are E = at for large tl and E = a for small tl . In
D-WAQ PART it is assumed that the dispersion process can be described in the same way
and that the dispersion coefficient can be fitted to a function with a power between these two
limits (0 < b < 1), which leads to:
For the horizontal direction the dispersion coefficient is defined as:
Dx,y = atb (6.4)
with a and b coefficients that follow from calibrating the model (Bent et al., 1991). The time
T
t is defined from t = 0, which is the time at which the given particle, for which dispersion is
calculated, is released.
Bed shear stress

AF
The bed shear stress at any location is calculated from:
τb =
ρg(u2 + v 2 )
(6.5)
C2
with:
τb bed shear stress [Pa]
ρ water density [kg/m3 ]
u flow velocity in x direction in the bottom layer [m/s]
v flow velocity in y direction in the bottom layer [m/s]
C Chézy coefficient [m1/2 /s]
DR
g gravitational acceleration [m/s2 ]
If the bed shear stress at any location is less than the critical shear stress for sedimentation,
a particle that comes into contact with the bottom at that location will remain attached to the
bottom (sedimentation). For sedimentation, D-WAQ PART creates an extra model layer for
sediment at the bed.
If the bed shear stress at any location is greater than the critical shear stress for sedimentation,
a particle that comes in contact with the bottom at that location will be reflected back into the
water column.
If the bed shear stress at any location is greater than the critical shear stress for erosion, all
deposited particles at that location (i.e. particles located in the extra bed-sediment layer) will
be returned to the water column instantaneously.
The settling velocity of each particle at time t is calculated from:

n 2π (t + ϕ)
vs = c × A0 + A1 sin (6.6)
T
where:
c local concentration of particles [kg/m3 ]
n exponent for adjusting concentration-dependent settling velocities
A0 the non-cyclic component of the settling velocity [m/s]
A1 the amplitude of a periodic sinusoidal variation in time [m/s]
T the period of the sinusoidal variation [hours]
ϕ the phase lag for the sinusoidal variation [hours]
Deltares 103 of 137

The factor in front of the square brackets enables you to define a concentration-dependent
settling velocity. If the exponent is set to n = 0, the settling velocity is simply a sinusoidal
function of time.
Remark:
⋄ For negative values of vs , the particles will move upwards towards the water surface.
6.3.3 Vertical dispersion

In a well-mixed, horizontally uniform flow, the vertical dispersion coefficient may be estimated
T
from the mixing length and the turbulent kinetic energy as:
1/4
√
cµ L k
Dz = (6.7)
σC
AF with:
Dz
cµ
vertical dispersion coefficient [m2 /s]
a constant; ≈ 0.09, calibrated for local equilibrium shear layers (Rodi, 1984)
L mixing length [m]
k turbulent kinetic energy
σC Prandtl-Schmidt number (= 0.7 for the transport of heat and salinity)
The Bakhmetev (1932) mixing length is estimated by:

p
L = κ(H − Z) (Z/H) with Z = −(z − ζ) (6.8)
DR
with:
κ von Kármán constant (= 0.41)
z the vertical co-ordinate
ζ the free surface elevation above the reference plan (at z = 0)
d the depth below the reference plane (bed level z = d)
H the total water depth, given by H = d + ζ
It should be noted that the Z co-ordinate is pointing downwards, i.e. Z = 0 is at the surface
and Z = H is at the bottom.
The turbulent kinetic energy is assumed to have a linear profile:
k = kb (Z/H) + ks (1 − Z/H) (6.9)
with:
kb turbulent kinetic energy near the bottom
ks turbulent kinetic energy near the surface
The empirical relationships for the turbulent kinetic energy at the bed and at the surface are
taken from the k -L turbulence model used in Delft3D-FLOW. This gives:
u∗ τb
kb = √ b = √
cµ cµ ρ0
u∗ τw (6.10)
ks = √ s = √
cµ cµ ρ0
with:
104 of 137 Deltares

Conceptual model
τb bottom shear stress [Pa]

τw wind stress [Pa]
ρ0 density of water [kg/m3 ]
The shear stresses are derived from the relations :
gρ0 |v|2
τb = 2
C2D (6.11)
τw = ρa Cd |w|2
T
with:
g gravity constant (= 9.81 m/s2 )
ρa density of air (= 1.25 kg/m3 )
C2D Chézy coefficient for 2D formulations (C2D = 50 m1/2 /s; unless specified oth-
AF
Cd
|v|
erwise in the sedimentation/erosion model)
wind drag coefficient of the free surface
advective velocity (computed from Delft3D-FLOW)
|w| wind velocity at 10 m above the water surface
The resulting vertical dispersion model (known as the k -L model) is obtained by substituting
the above expressions into the expression for the vertical dispersion coefficient to obtain:
κ √ q
Dz (z ′ ) = H(1 − z ′ ) z ′ u2∗b z ′ + u2∗s (1 − z ′ ) (6.12)
σC
DR
where z ′ = Z/H
Figure 6.1 shows an example of the k -L vertical dispersion coefficient (and its gradient that
causes particle drift) for a wind speed of 7.5 m/s.
Figure 6.1: Vertical disperison coefficient [m2 /s] and its gradient (dashed line) as a func-
tion of relative depth for the k -L turbulence model, at a wind speed of 7.5 m/s
and a Chézy coefficient of 50.0.
Deltares 105 of 137

D-WAQ PART does not use space-dependent dispersion coefficients as this would lead to
additional stochastic drift terms. From Figure 6.1, it can be seen that particles would drift
towards the bottom and the surface and would accumulate there. In order to prevent such non-
physical behaviour, the depth-dependency of the vertical dispersion coefficient is eliminated
by depth-averaging. D-WAQ PART allows linear scaling of the depth-averaged dispersion
coefficient to allow for a reduction in vertical mixing due to stratification in the 3D models.
The depth-averaged relation for Dz is obtained by integrating the above expression to obtain:
Z z ′ =1
κH
T
q
Dz = u2∗b z ′2 (1 − z ′ )2 + u2∗s z ′ (1 − z ′ )3 dz ′
σC ′
"zZ=0′ #
z =1 Z z ′ =1
κH
q q
2
≈ ′
u2∗b z ′2 (1 − z ′ ) dz + u2∗s z ′ (1 − z ′ )3 dz ′ (6.13)
σC z ′ =0 z ′ =0
AF =
κH h u∗b
σC 6
u∗ i
+π s
16
6.4 3 Dimensional flow field

D-WAQ PART is a 3D-model. A 3D hydrodynamic flow field is obtained from a 3D Delft3D-
FLOW hydrodynamic simulation (operation in mode 1) or by superimposing a logarithmic ve-
locity profile over depth for the bottom shear on a 2DH Delft3D-FLOW (operation in mode 2)
simulation and a parabolic profile over depth for wind. In this section the velocity profiles for 2
DH calculations are discussed:
DR
Logarithmic profile for bottom shear

The logarithmic profile for bottom shear is given by:

H −Z
vz = C ln +1 ×v (6.14)
Z0
where C is:
−1
Z0 H
C= 1+ ln +1 −1 (6.15)
H Z0
with:
vz vertical profile of the horizontal velocity
Z depth co-ordinate (from 0 (surface) to H (bottom))
H water depth
Z0 roughness length
v depth-average velocity
The roughness length is of the order 2 cm. The depth-averaged velocity is obtained from a
2DH-version hydrodynamic program (Delft3D-FLOW).
Parabolic profile for wind

The parabolic profile for wind is given by:

Z Z
vz = α 3 1 − −2 1− W (6.16)
H H
with:
106 of 137 Deltares

Conceptual model
W wind speed at 10 m above sea level

α wind drag coefficient
The wind-drag coefficient is typically in the order of 0.03. The integral over depth is zero
and is superimposed on the normal velocity profile which already includes the gross average
influence of the wind. This gross wind effect on the velocity field must have been incorporated
in the 2DH-hydrodynamic velocity field.
6.5 Oil spill module
T
The D-WAQ PART Oil Spill model calculates the transport, spreading, evaporation and dis-
persion of an oil patch. Transport is either 2D or 3D. Oil can either be floating on the water
surface, or it can be transported as a dispersed phase in the water column. Oil dispersion,
also referred to as entrainment, is generated by breaking waves (due by wind) according to
a formulation of Delvigne et al. (1986). Evaporation is based on a simple first order decay
AF process or on formulations by Fingas (?, ?).
Release of oil
Oil can be spilled as an instantaneous release or as a continuous release. The radius of an
instantaneous release may follow from the standard input menu of the release, but may also
be specified by the following formulation (Fay and Hoult, 1971):
 1/12
ρw −ρ0
k22 V05 g ρw
R0 =   (6.17)
k1 νw2
DR
with:
V0 initial volume of the oil spill [m3 ]
ρ0 oil density [kg/m3 ]
ρw density of water [kg/m3 ]
g gravity constant
νw kinematic viscosity of water [10−6 m2 /s]
k1 , k2 constants of Fay (1.14 and 1.45)
You should specify the oil density ρ0 as input. The radius describes a patch after the so-called
‘gravity inertial phase’ which lasts in the order of 5 minutes for most spills. The thickness of the
floating patch is then usually less than 1 mm. It is not recommended to use this description
for continuous releases, but to specify the radius of the release according to the information
that is available.
Wind induced advection of surface oil

The advection of surface floating oil is subject to wind effects. This is widely published in liter-
ature. For example, Labelle and Johnson (1993) carried out simulations of oil-spill trajectories
for which each trajectory was constructed using vector addition of the current field and 3.5
percent of the instantaneous wind. A drift angle was computed as a function of wind speed
(inversely related to wind speed). In a review by the Task Committee on Modelling of Oil Spills
of the Water Resources Engineering Division (1996), it has been reported that the majority
of spill models use a simplified linear superposition technique to approximate spill motion.
The currents induced by winds and waves are normally lumped together and represented by
an empirically based drift factor and deflection angle dependent on the local wind speed and
direction. Drift speeds typically vary from 2.5–4.4 % of the wind speed. The deflection an-
gles vary between 0 and 25 degrees to the right/left of the wind direction (northern/southern
Deltares 107 of 137

hemisphere). The most detailed description of drift of oils is given by Yousseff and Spaulding
(1993). Here also the drift speed is specified as typically between 2.5 and 4 % of the wind
speed with a mean value of 3.5 %, and deflection angles between 0 and 25 degrees, with a
mean of 15 degrees. Yousseff and Spaulding (1993) deal in detail with the effects of waves
on the transport.
A representation of the effect of wind in D-WAQ PART is implemented for oil. For a reasonable
behaviour of surface floating oil under windy conditions, it is essential for the wind to affect the
advection of the surface floating oil. The most important effect is drift as a percentage of the
wind speed. The relationship that describes this effect is specified as
T
Cwd (Vw − Vf ) (6.18)
with Cwd the wind drag, Vw the wind speed and Vf the current speed. This relationship is
only applied to surface floating oil because the dispersed oil will be transported correctly by
AF
the currents of the hydrodynamic model and not be directly influenced by wind.
Another factor that affects the transport of the oil is governed by the aforementioned deflection
angle, which is an angle between the wind direction and oil advection. This is essentially
caused by the fact that the effects of the waves is under the influence of Coriolis and that the
wave induced transport is at an angle of the wind. This angle is an empirical parameter. This
parameter is chosen to be a constant and it is therefore assumed that the angle does not
depend on the wind speed. The deflection angle will depend on the latitude. The deflection
angle is included in the D-WAQ PART model as an additional parameter. The deflection
angle would therefore essentially be a calibration parameter, depending on latitude and wind
strength.
DR
The drift of the surface oil is implemented in the 3-dimensional mode. When the flow is in a
2D mode, the advection by wind drift at the surface is included in the vertical profile that is
derived in D-WAQ PART (see section 6.4), thus an additional wind drift is not required.
Evaporation of oil
Evaporation of floating oil is implemented as a first order decay process, the rate is either
explicitly specified or it is calculated according to the formulations by Fingas (?, ?). Decay
can also be specified by a decay constant as usual in the menu of process parameters (this
can be done with a time-series), or by defining a fixed fraction of oil that decays each day.
Fingas proposed to describe the time-dependent evaporation using a natural log or a square
root function, with a temperature dependency (?):
For oils (most oils and petroleum products) that follow a logarithmic equation:
Percentage evaporated = [.165(%D) + .045(T − 15)] ln(t) (6.19)
For oils like diesel fuel that follow a square root equation:
√
Percentage evaporated = [.0254(%D) + .01(T − 15)] t (6.20)
where:
%D is the percentage (by weight) that evaporates at 180o C
T the water temperature in o C
t the age (time sicne release into the water) in minutes.
108 of 137 Deltares

Conceptual model
Within the D-WAQ PART model the formulation with the logarithm or the square root is se-
lected via the water temperature: A positive temperature is used to select the logarithm and a
negative temperature to select the square root formula.
Thus, with the evaporation description these two parameters will need to be supplied by the
user:
• the (water) temperature
• the mass percentage oil evaporated at 180o C
T
It is known that oil contains fractions that do not evaporate. For example, Reed (1989) adopted
in his model a mass transfer coefficient that uses the molecular weight of the volatile fraction
of the oil spill.
In the oil module of PART, a volatile fraction has been introduced, albeit simplified compared
AF
with the implementation of Reed (1989). It is assumed that the volatile fraction Fvol evap-
orates as a first order process (i.e. exponential), and that the non-volatile fraction does not
evaporate at all. This is achieved by the introduction of the following:

dFv Fvol − Fv
=− k (6.21)
dt 1 − Fv
Where Fv is the evaporated fraction and k the evaporation rate constant. The numerator in
the equation should always be positive and if Fvol − Fv < 0 then the evaporation is set to
zero.
DR
The oil viscosity will change during the evaporation process. In the literature, descriptions of
the dependency of the viscosity as a function of the evaporated fraction exist and is generally
given as:
η = η0 e(Cv Fv ) (6.22)
This equation is given for the dynamic viscosity. Assuming that the density is approximately
constant throughout the simulation, then the same function can be used to describe the kine-
matic viscosity. Reed (1989) states that the value of Cv in his model is equal to 1 for gasoline,
kerosene and light diesel fuel and 10 for other petroleum products. In the D-WAQ PART
model, this is implemented by assuming that C = 1 is used for light oils (kinematic viscosity
less than 500 cSt) and C = 10 for the heavy oils (viscosity greater than 500 cSt).
Dispersion (entrainment) of oil

Dispersion of floating oil, or entrainment of oil in water, is implemented as a zero order decay
process, i.e. the entrainment rate is independent of the floating oil concentration. The disper-
sion rate of oil depends only on the wave energy that is dissipated by the patch, and the type
of oil.
The dispersion rate can be calculated in three ways:

⋄ As a constant fraction per day (so independent of wind and waves).
⋄ According to the formulation by Delvigne and Sweeney, as shown below, where the thresh-
old for the wind velocity where white capping occurs is 5 m/s.
⋄ According to the formulation by Delvigne and Sweeney, but with a wind velocity threshold
that is user-defined.
Deltares 109 of 137

In the second and third option, the dispersion rate Q [kg/(m2 /s)] of oil is given (see Delvigne
and Sweeney (1988); NOAA (1994); Delvigne and Hulsen (1994)) by:
Z δmax
Q= Q(δ) dδ
deltamin
Q(δ) = C ′′ De0.57 Fwc N (δ)δ 3
N (δ) = N0 δ −2.3
√
De = 0.0034ρw gH0 / 2
T
0.243Uw2 (6.23)
H0 =
g
fw
Fwc =
tp
AF tp = 8.13
Uw
g
fw = max(0, 0.032(Uw − 5.0))
with:
Q dispersion rate [kg/(m2 /s)]
Q(δ) dispersion rate per unit of diameter for droplets of diameter d [kg/(m2 /s)]
δ oil droplet diameter [m]
δmin minimum oil droplet diameter [m]
deltamax maximum oil droplet diameter [m]
DR
C ′′ oil constant (calibration parameter, depending on kind of oil)

N (δ) oil particle size distribution function
N0 normalisation constant distribution function
De dissipation of wave energy per unit surface area [J/m2 ]
Fwc number of waves that break per wave period [-]
tp peak wave period [s]
Uw wind speed [m/s]
fw fraction of sea covered by white caps [-]
Here, white capping is formulated according to Holthuijsen and Herbers (1986) with the initial
wind speed for white capping equal to 5 m/s. The minimum droplet size δmin can be taken zero
as a good approximation. The crux is knowledge of parameters δmax , N0 and the calibration
constant C ′′ .
According to NOAA (1994), after resurfacing of particles back into the oil slick, δmax can be
taken equal to 70 µm. In this approximation, it is assumed that after each breaking wave a
quasi-steady state distribution of droplets results, i.e. the resurfacing of particles back into the
oil slick due to buoyancy goes fast compared with the dissipation of wave energy by the oil
slick. Defining a new calibration constant C0 the following expression results for Q:
Q = 5.08 · 10−8 C0 Scov De0.57 Fwc (6.24)
with
Q dispersion rate [kg/(m2 /s)]
C0 oil constant (calibration parameter)
Scov proportion of the sea surface covered by the oil in the relevant area
De dissipation of wave energy per unit surface area [J/m2 ]
Fwc number of waves that break per wave period [-]
110 of 137 Deltares

Conceptual model
Relationship viscosity and Co

10000
log(Co) = -0.0658*log(visc) + 3.2618
1000
log(Co)=-1.1951*log(visc)+5.6456
100
Co
T
10
Measured(Delv&Hulsen)
Low viscosity (measured)

AF 1
1
High viscosity (measured)
10 100
viscosity (cSt)
1000 10000
Figure 6.2: Relationship between viscosity and the dispersion parameter C0 (Delvigne
and Hulsen, 1994)
The calibration parameter C0 depends on the kind of oil: oil with a high viscosity disperses
hardly for a wind speed of 10 m/s whereas oil with a low viscosity disperses fast for such a
wind speed.
DR
An order of magnitude of C0 follows from Delvigne and Hulsen (1994) (Table 2 and Figure 4 of
that publication). Since Delvigne and Hulsen do not take into account a steady state assump-
tion their values can only be a first estimate for D-WAQ PART. Estimates of the dispersion
constant C0 are:
C0 ≈ 2000 for Ekofisk with standard oil viscosity ν = 8 cSt (at 20 ◦ C)
C0 ≈ 50 for Heavy Fuel Oil with standard oil viscosity ν = 3000 cSt (at 20 ◦ C)
Since viscosity is temperature dependent C0 depends both on oil type and on temperature.
Temperature dependence may be neglected for low viscous oils with ν < 100 cSt (Delvigne
and Hulsen, 1994).
Delvigne and Hulsen (1994) have shown that there is a relationship between the C0 and
the oil viscosity. Since viscosity is an oil characteristic and used as input of the model, this
relationship is included in the model. The relationship used in the modified PART(Oil) model
is based on the data published by Delvigne and Hulsen (1994). The data was examined and
a log-log plot revealed that this relationship can be specified by two equations (Figure 6.2).
The equations used and the goodness of fit with the data are indicated in the figure.
In order to calculate the entrainment flux, some representation of surface area for each surface
oil particle is required. In the oil module an area for each particle is derived, based on the mass
of surface oil associated with this particle (WL | Delft Hydraulics, 2003). This is calculated
using an assumed oil layer thickness. Validation with the oil budget model ADIOS has resulted
in an optimal value of 0.00005 m. This value of 0.05 mm (50 µm) is also similar to the value
of 70 µm generally used to define the oil droplets that remain in suspension.
Deltares 111 of 137

Emulsification
Water-in-oil (w/o) emulsions form a viscous cream, or floating, coherent semi-solid lumps,
often called chocolate mousse. The process depends on oil composition. The formation
of w/o emulsions reaches a maximum between 10 and 100 hours (Wheeler, 1978). The
emulsification process itself is relatively rapid. It has been reported that emulsification takes
place fully in the laboratory in 0.1–3 hours (Fingas et al., 1999).
Emulsification is virtually an irreversible process (Wheeler, 1978) and changes the liquid to a
heavy, semi-solid material (Fingas and Fieldhouse, 1996) with a high viscosity. Viscosity can
be as high as 4 · 106 cP (Bos, 1980).
T
The emulsification process is implemented in D-WAQ PART following the algorithm presented
by Mackay and others (Mackay et al., 1980; Zagorski and Mackay, 1982), in which the rate of
water uptake F̃wc is given by:
AF F̃wc

= C1 (Uw + 1) 1 −
Fwc
C2

2
(6.25)
with Uw the wind speed, Fwc the water content and C1 and C2 model parameters. C1 is
given as 2 · 10−6 for emulsifying oils and 0 for others, whilst C2 is a constant controlling the
maximum water content and is suggested to be 0.25 for home heating oil and 0.7 for crude
and heavy fuel oil (Reed, 1989). In several other publications, however, the maximum water
content for these heavy oils is said to reach values of 75–80 % (for example, Huang (1983)).
The switch of C2 between 0.25 and 0.75 is made at a viscosity of 500 cSt, whilst the value
DR
of C1 (0 or 2 · 10−6 ) is selected by the user and thus dependent on the emulsion formation
tendency of the oil in question.
The main effect of emulsification (increase in the water content) in the model is the change in
viscosity and is given as:
µ

2.5Fwc
=e 1.0−C3 Fwc
(6.26)
µ0
As with the change of viscosity as a function of the evaporated fraction, it is assumed that the
density does not change significantly, compared with the changes in viscosity and is assumed
constant. Thus the implementation in the PART(Oil) model uses the kinematic viscosity in the
aforementioned equation instead of the dynamic viscosity. The constant C3 is suggested to
be 0.65 (Reed, 1989).
Emulsification does not only affect viscosity (and therefore the dispersion (entrainment) pro-
cess) but also evaporation. According to Fingas (1994), the effect of emulsification is that the
viscosity rises two to three orders-of-magnitude, the spreading rate decreases by a similar
value and evaporation nearly ceases.
In order to achieve a link between the emulsification and evaporation in PART(Oil), the water
content of the emulsion is used to reduce the evaporation rate. In the implementation in PART,
it is assumed that the evaporation ceases when the water content has reached its maximum.
This maximum water content is oil type dependent. The evaporation rate is adapted when
emulsification occurs by reducing the volatile fraction. This is implemented as follows:
C2 − Fwc
Few = Fvol (6.27)
C2
112 of 137 Deltares

Conceptual model
Where Fvol is the volatile fraction of the spilled oil and Few the adapted volatile fraction,
replacing the volatile fraction Fvol in Equation (6.21). Thus, when the water content reaches
its maximum C2 , the adapted volatile fraction reduces to zero and the evaporation halts.
The onset of emulsification can be delayed until evaporation causes the oil characteristics
to reach the criteria for emulsification. This delay has been implemented in PART(Oil) by
introducing a fraction evaporated oil at which emulsification is initiated (Ev ).
Oil density
T
The density of the oil is affected by evaporation and emulsification. Density is also a function
of temperature. In PART, the ADIOS formulation is used, without the dependency of the
evaporated fraction and temperature. Hence, in PART, the density can only change when
the oil emulsifies. The density is then a linear interpolation of the densities of water and oil,
according to their relative content:
AF ρem = Fw ρw + (1 − Fw ) ρoil (6.28)
where
ρem Density emulsion [kg/m3 ]
Fw Water fraction
ρw Water density [kg/m3 ]
ρoil Oil density [kg/m3 ]
Sticking of oil
DR
Oil may stick to land and cause damage to ecological sites. In order to simulate the process of
oil sticking to land or to the bed of the water column, a sticking probability must be specified.
A particle may come into contact with land due to wind effects or horizontal dispersion and it
may come into contact with the bed due to vertical dispersion or settling. The particle sticks
to the land or bed if a randomly chosen number between 0 and 1 is smaller than the sticking
probability given.
Settling of oil particles

Dispersed oil may settle. However, this velocity is generally negative since oil particles are
lighter than water. Most oil and refined products are less dense than water, which means they
will float when initially spilled. Experience has shown that oil slicks that initially float will remain
buoyant even after weathering. The two circumstances under which oil has been noted to sink
occur when the oil is mixed with mineral sediment in the water column or when it is burned
in-situ, creating a residue that may be of high density.
Weathering of oil
In addition to the processes mentioned earlier, other weathering processes may affect the
presence of oil. Processes such as oxidation, bacteriological decay can in D-WAQ PART be
simulated by means of a first order decay, for which a decay parameter can be specified.
Deltares 113 of 137

6.6 Data requirements for the conceptual model

A list of data that enter the conceptual model and pertain to the physical system that is being
modelled will be described in this section. Assumptions are made about the sensitivity of
the model results for these parameters. These assumptions are general remarks and their
importance strongly depends on the accuracy required by you for giving you enough insight
in the particular water quality problem at hand.
1 Hydrodynamic database of volumes and flows

Source supplied by the coupling module after a Delft3D-FLOW run
T
Meaning determines the advective transport and the amount of water avail-
able for dilution of the wasted substance
Sensitivity should be accurate, especially for the calibration of dye experi-
ments
2 Wind drag coefficient
AF Source
Meaning
user-defined
parabolic wind profile over depth (2D flow field); drag on surface
oil (Oil module).
Sensitivity anisotropy of a patch for a dye experiment depends on this pa-
rameter (range 1 %–5 %). Experience at Delft Hydraulics learns
that a value of 1 % is mostly used for D-WAQ PART applications
rather than the 3 % recommended by literature. This may be
due to the D-WAQ PART algorithm (parabolic profile and depth-
averaged vertical dispersion coefficient).
3 Wind field
DR
Source user-defined
Meaning describes parabolic wind profile for 2D FLOW velocity field, used
in the 2DH approach. If the depth-averaged vertical dispersion
option is selected, then the wind is also used to calculate the ver-
tical dispersion. For the oil module, the wind is used to calculate
the entrainment flux of the oil into water.
Sensitivity the average direction and wind speed should be a reasonable
estimate
4 Deflection angle (3D)
Source user-defined (literature value), depending on latitude
Meaning describes the angle between the wind direction and oil advec-
tion.
Sensitivity The deflection angle is generally between 0 and 25 ◦ rotating to
the right on the northern hemisphere and to the left on the south-
ern hemisphere.
5 Roughness length (2DH model)
Source user-defined (literature value)
Meaning describes logarithmic profile due to bottom shear for 3D velocity
field
Sensitivity a value of 2 cm is recommended
6 Dispersion parameters a and b for horizontal dispersion
Source user-defined
Meaning determines the increase of the horizontal dispersion coefficient
with time.
Sensitivity the parameter a is most important for the short-time range whereas
b is most important for the long-time range. The value of the
parameters depends on the amount of detail (turbulence) not in-
114 of 137 Deltares

Conceptual model
corporated in the hydrodynamics. The parameters should follow

from calibration or from an expert opinion.
7 Amount of released material
Source user-defined
Meaning an instantaneous release (mass) or a continuous release (re-
lease rates and concentrations or, alternatively, mass rates).
Sensitivity quantitative results depend linearly on the accuracy of these in-
put data
8 Time and location of released material
T
Source user-defined
Meaning -
Sensitivity should be accurately known
9 Constants oil spill module
AF
Source user-defined, including oil-viscosity, from which the dispersion
(entrainment) parameter is derived, see Delvigne and Hulsen
(1994). A stickyness probability is important for sticking of oil
to land or the bed. All oil model parameters are related to the oil
characteristics. Oil characteristics are available from oil databases.
DR
Deltares 115 of 137

T
AF
DR
116 of 137 Deltares

7 Algorithmic implementation
7.1 Introduction
For each particle, transport is split in a part for advection and a part for wind and horizontal and
vertical dispersion. Dispersion and advection of clouds of particles are treated with different
numerical schemes. These schemes are described in this section.
7.2 Advection schemes
T
The advective part is solved with an analytical integration procedure that integrates exactly
a linearly interpolated hydrodynamic velocity field. The advective velocity of a particle is ob-
tained by linear interpolation of the velocities at the borders of the computational cells of the
grid of the hydrodynamic simulation. A vertical velocity profile is added to the interpolated
velocity field. The path of a particle in a computational grid-cell is determined by analyti-
AF cal integration along the velocity field. This procedure leads to mass-conservation of water.
The analytical integration method automatically deals with closed (land) boundaries correctly,
since velocities vanish asymptotically at the borders. Particles that pass open (water) bound-
aries are taken out of the calculation.
The analytical integration procedure is given by:

Z ∆t
dx
x(t + ∆t) = x(t) + dt (7.1)
0 dt
with:
DR
x is the distance
t is the time
dx
dt
is the velocity in x-direction
x = x(s(t))
dx ds
= (αx x(s) + βx )
dt dt
Q+ − Q− (7.2)
αx = C(σ)
V
Q+/−
βx = C(σ)
V
with:
s(t) is the streamline distance function
αx(s) + β the linearly interpolated velocity in the grid along the stream line
Q+/− is the flow though the upwind or downwind surface of the grid cell
V is the volume of the grid cell
C(σ) is the distance between the velocity points in the hydrodynamic grid
Q+/−
V
is 1/travel time of the grid cell with this Q+/−
For the implemented numerical integration, the path acceleration is neglected:
ds
=1 (7.3)
dt
This flow field is exactly related to the hydrodynamic flow field (Delft3D-FLOW).
Deltares 117 of 137

The flow field is conservative if:
∂
(αx x + βx ) = 0 (7.4)
∂x
and for coupling to Delft3D-FLOW this condition is satisfied.
7.3 Dispersion schemes

The dispersive part is solved with an Euler-type numerical scheme. In the horizontal direction
also additional advection due to wind (in 2DH mode) is handled with this numerical scheme.
T
In vertical direction a displacement due to settling can be added to the vertical displacement
due to dispersion. Boundary conditions are handled simultaneously for these two schemes
and are purely reflective.
Dispersion in both the horizontal and the vertical direction is based on a white noise process
AF with a uniform probability density function with zero mean ranging between -1 and +1. The
maximum displacement of each particle for each time-step is then:
√
∆S = 6D∆t (7.5)
It is noted here that none of the three dispersion coefficients is allowed to be 0, since this will
reduce the numbers of degrees of freedom which will affect the specification of the displace-
ment.
DR
118 of 137 Deltares

References
Bakhmetev, B. A., 1932. Hydraulics of open channels,. Eng. Soc. Monograph. McGraw-Hill.
Bent, E., L. Postma, A. Roelfzema and R. Stive, 1991. “Hydrodynamic and dispersion mod-
elling of Swansea Bay.” Environmental Hydraulics 1: 865–870.
Boogaard, H. van den, M. Hoogkamer and A. Heemink, 1993. “Parameter identification in

particle models.” Stochastic Hydrology & Hydraulics 7: 109–130.
Bos, J., 1980. Gedrag van olie op zee. Tech. rep., Rijkswaterstaat, Directie Noordzee.
T
Csanady, G., 1973. Turbulent Diffusion in the Environment. Reidel, Dordrecht.
Deltares, 2013. Delft3D-FLOW User Manual. Deltares, 3.14 ed.
Deltares, 2023. D-Water Quality User Manual. Deltares, 5.06 ed.

AF
Delvigne, G. and L. Hulsen, 1994. “Simplified laboratory measurements of oil dispersion
coefficient: application in computations of natural oil dispersion.” In Proc. 17th Arctic &
Marine Oil Spill Program, Vancouver, pages 173–187.
Delvigne, G., J. Roelvink and C. Sweeney, 1986. Research on vertical turbulent dispersion of
oil droplets and oiled particles. OCS Study MMS 86-0029, US Department of the Interior,
Anchorage.
Delvigne, G. and C. Sweeney, 1988. “Natural dispersion of oil.” Oil & Chemical Pollution 4:
281–310.
DR
Fay, J. and D. Hoult, 1971. Physical processes in the spread of oil on a water surface. Report
DOT-CG-01 381-A, U.S. Coast Guard, Washington D.C.
Fingas, M., 1994. “Chemistry of oil and modelling of spills.” Journal of Advances in Marine
Technology Conference 11: 41–63.
Fingas, M. and B. Fieldhouse, 1996. “Oil spill behaviour and modelling.” In Paper presented
at Eco-Informa ’96, Lake Buena Vista, Florida.
Fingas, M., B. Filedhouse and J. Mullin, 1999. “Water-in-oil emulsions results of formation
studies and applicability to oil spill modeling.” Spill Science & Technology 5 (1): 81–99.
GPP UM, 2013. Delft3D-GPP User Manual. Deltares, 2.14 ed.
Holthuijsen, L. and T. Herbers, 1986. “Statistics of breaking waves observed as whitecaps in

the open sea.” Journal of Physical Oceanography 16 (2): 290–297.
Huang, J., 1983. “A review of the state-of-the-art of oil spill fate/behaviour models.” In 1983
Oil Spill Conference.
Krone, R., 1962. Flume studies of transport of sediment in estuarial shoaling processes
(Final report). Tech. rep., University of California, Hydraulics Engineering and Sanitary
Engineering Laboratory, Berkeley, USA.
Labelle, R. and W. Johnson, 1993. “Stochastic oil spill analysis for Cook Inlet/Shelikof Strait.”
In Proc. 16th Arctic and Marine Oil Spill Techn. Seminar, Environment Canada, Ottawa,
pages 573–584.
Mackay, D., S. Paterson and K. Trudel, 1980. A mathematical model of oil spill behavior on
water with natural and chemical dispersion. Tech. Rep. EPS-3-EC-77-19.
Deltares 119 of 137

NOAA, 1994. ADIOS User’s manual, Version 1.1. Seattle, Washington, 1.1 ed.
Partheniades, E., 1965. “Erosion and Deposition of Cohesive Soils.” Journal of the Hydraulics
Division, ASCE 91 (HY 1): 105–139. January.
Reed, M., 1989. “The Physical fates Component of the Natural Resource Damage Assess-
ment Model System.” Oil & Chemical Pollution 5: 99–123.
Rodi, W., 1984. “Turbulence models and their application in Hydraulics, State-of-the-art paper
article sur l’etat de connaissance.” IAHR Paper presented by the IAHR-Section on Funda-
mentals of Division II: Experimental and Mathematical Fluid Dynamics, The Netherlands.
T
Rubinstein, R., 1981. Simulation and the Monte Carlo Method. Wiley, New York.
Task Committee on Modelling of Oil Spills of the Water Resources Engineering Division, 1996.
“State-of-the-art review of modelling transport and fate of oil spills.” Journal of Hydraulic
AF
Engineering ASCE: 594–609.
WAVE UM, 2013. Delft3D-WAVE User Manual. Deltares, 3.03 ed.
Wheeler, R., 1978. The Fate of Petroleum in the Marine Environment,. Special report, Exxon
Production Research Company.
WL | Delft Hydraulics, 1993. Parameter Identification in Delft3D-PART. Tech. Rep. T880, WL

| Delft Hydraulics, Delft, The Netherlands. (R.J. Vos and H.F.P. van den Boogaard).
WL | Delft Hydraulics, 2003. Modelling transport and fate of oil in the marine environment.
Report Z3291, WL | Delft Hydraulics, Delft, The Netherlands. (Kleissen, F.M.).
DR
Yousseff, M. and M. Spaulding, 1993. “Drift current under the action of wind and waves.” In
Proc. 16th Arctic and Marine Oil Spill Techn. Seminar, Environment Canada, Ottawa, pages
587–615.
Zagorski, W. and D. Mackay, 1982. “Water in oil emulsions: a stability hypothesis.” In Proc.
5th Annual Artic Marine Oilspill Program Technical Seminar, Environment Canada, Ottawa.
120 of 137 Deltares

A Input files D-WAQ PART
A.1 Introduction
The particle model gets its information from the hydrodynamic model (Delft3D-FLOW) and a
user input file. An input file is normally generated by the Graphical User Interface (GUI) but
may also be edited directly using any text editor. However, you should always be careful when
editing the input file directly, since inconsistencies in the input that would not be possible
when using the GUI, may lead to unexpected results, without generating appropriate error
messages.
T
This appendix describes the main ASCII input file for D-WAQ PART. In addition to this input
file, PART also requires the file <runid.mdp> (see section 3.4.3) and the hydrodynamics files
generated by the coupling program.
AF If the oil module is selected and initial condition is specified, an external initial condition file
<∗.ini> will be required. The format of this file is in Section A.6.
A.2 General set-up of the input file

Formats
The user input file is free formatted. You may insert information anywhere on lines and may
insert comment lines or comments on data lines in the following way:
⋄ All information after a semicolon (;) is treated as comment, the semicolon included. (This
DR
holds also for semicolons in character fields!)

⋄ Blank lines are skipped and may be used freely.
⋄ Character strings must be supplied in quotes.
The FORTRAN language may (sometimes) have problems with two subsequent leading zeros
in a number, so it is advised not to use multiple leading zeros in the free formatted input.
Errors in input
The input program checks as much as possible for wrong input. It is advised to review the
input and run report file carefully. Error messages are not given on screen but only given in
the report file, so always check the report file!
Specification of time functions

Time-functions are always specified by a number of breakpoints (days, hours, min, sec), fol-
lowed by a value for that breakpoint. The program requires the first breakpoint of the time-
series to be the same as the start of the simulation and the last breakpoint to be the same as
the end of the simulation.
By default, the program carries out a linear interpolation between the break points for all
specified time-functions. The program can mimic block functions by specifying two subse-
quent records with equal time and distinct values. For continuous releases you can select a
block function or linear interpolation of the time-function.
Deltares 121 of 137

Information in the input file

In this section the input file for model type 2 — oil simulation — is defined (2.5 DH or 3D
modes). In the explanation of the input file the following symbols are used:
- this item must be specified on a new line.

∗ this item must be specified in the input line on the same line as the last line
starting with a ’-’).
∆ indicates that this input is optional, or its use is only required when some other
variable indicates this (since it is > 0).
T
ASCII Input file (definition of):
'V3.66.00 ' ; please, don't change this line
'Tutorial \DPART'
'Friesian Tidal Inlet model'
AF 'Ekofisk and HFO release'
'T0: 1990.08.05 00:00:00 (scu =
- Run identification.
ls)'
The run identification consists of 5 strings of at most 40 characters. The first line is
required and identifies the PART version number that is used. This line should not be
changed since the model will not run when the version number in the input file is not con-
sistent with that of the model. The subsequent 4 lines are for information purposes and
are used on the graphical output. Do not put any quotes in the strings since this will lead
to blocking errors that are difficult to trace back.
; Hydrodynamics file (*.hyd)
'com-tut_fti_part.hyd'; name of *hyd file
DR
;
- Name of Delft3D <∗.hyd> file (see section 3.4).
This filename consists of at most 80 characters. The file needs to be located in the working
directory. The file is generated by the coupling program and contains all the input and
output filenames, including the ASCII input file discussed here. It should be noted that
the information in the <∗.hyd> file is used by the PART-GUI. When the GUI is exited the
<filename.dat> file is generated. This file includes the filenames and locations of the
hydrodynamics that will be used. Thus when executing D-WAQ PART, the information
from the <filename.dat> file is used, and not the <∗.hyd> file.
; Numerical parameters
; Type of model: 1 - tracer (basic module)
; 2 - (obsolete option)
; 4 - oil model
4 0 0 0 ; type of model tracks(0/1) extra_output(0/1) sed/erosion(0/1)
- An integer specifying the model-type (or ’mode’) — see section 3.4.
∗ An integer specifying whether particle tracks for all particles are generated in the out-
put. A value of 0 indicates no particle tracks output. A value of 1 will generate output
for each calculation time step. A value higher than 1 will be used as a multiplier for
the time step at which the tracks are written (this option is not supported by the user
interface). It is recommended to keep the number of particles in the model small (e.g.
less than 1 000) or use a high multiplier, because the output file can become very large
if a large number of particles is used. When the number of particles is greater than
1 000, a warning is generated.
∗ An integer that is no longer used, but is required in the input file.
∗ An integer specifying whether sedimentation/erosion processes are switched on (1) or
off (0).
122 of 137 Deltares

Input files D-WAQ PART
1 900 ; num. scheme time step(s) (0=flow time step)

- An integer indicating the numerical scheme for advection of the particles (option 1 is the
only option available).
∗ An integer indicating the time-step in seconds. A value of ‘0’ means that the time-step
will be the same as that of the hydrodynamic database.
1 0.10 0 ; vert. disp option(*) scale disp_coeff.(m2/s)
; (*) 0=constant 1=depth averaged
- An integer indicating the numerical scheme for vertical dispersion of the particles.
For the recommended option 0, a constant vertical dispersion parameter is used. For
T
option 1 this parameter follows from an algebraic k − L model.
∗ A real indicating a scale factor for vertical dispersion.
∗ A real indicating a vertical dispersion constant when dispersion option ‘0’ was used.
When option ‘1’ is used this value is not used.
AF;
6
'Ekofisk
'Ekofisk
2
'
Substances
; no of substances
disp'
no of oil fractions
'Ekofisk stick'
'Heavy Fuel Oil'
'Heavy Fuel Oil disp'
'Heavy Fuel Oil stick'
- An integer specifying the total number of substances.
∗ An integer specifying the number of oil substances
DR
- Substance name.
For each substance, a string of 20 characters (maximum) must be given. For the oil spill
module, for each oil fraction, specify 3 names in the following order:
1 Name of the floating Oil, e.g. ‘Ekofisk’;
2 Name of the dispersed Oil, e.g. ‘Ekofisk dispersed’;
3 Name of the Sticking Oil, e.g. ‘Ekofisk sticking’.
Because of this, we advise to use a substance name with a maximum of 10 characters
each, so that the sub-types dispersed and sticking can be added.
; Particles
20000 ;number of particles
- Number of particles.
You may select the total number of particles to be used. More particles results in a more
accurate solution of the mathematical equations (not necessarily in a better simulation)
and a better resolution of the concentration, but in a longer computation time. The maxi-
mum number of particles is limited by the memory of your computer.
Please note that this number is not necessarily the total number of particles used in the
simulation. This is because the total number of particles in the simulation will be the sum
of all percentages of particles that is specified for every release (continuous and instan-
taneous) multiplied by the specified number of particles. Thus the number entered here
represents a total release of 100 % of the particles (see instantaneous and continuous
releases).
; Physical parameters
0.002 ; roughness [m]
0.07 ; a coeff. in hor. diff. D = a*t^b
0.7 ; b coeff. in hor. diff. D = a*t^b
3 ; wind_drag (%)
1024 ; density of water (kg/m3)
Deltares 123 of 137

- Roughness length [m].

The roughness length parameter is one of the calibration constants (see the report con-
cerning the principles of the model). It is advised to take something like 0.02 m. This
constant is only used when D-WAQ PART runs in 2DH mode, but must also be supplied
for 3D calculations.
- Isotropic dispersion coefficients [m2 /s].
This constant a is also subject to calibration. See section 6.3 for details.
- Exponent for horizontal dispersion coefficients.
This constant b is also subject to calibration. See section 6.3 for details.
- Wind drag coefficient [%].
T
This constant represents the velocity of the top most water layer as a percentage of the
wind velocity (e.g. 3.0 means 3 %). For the tracer module, this constant is only used when
D-WAQ PART runs in 2DH mode, even though it must also be supplied for 3D calculations.
The value is also used by the oil module for the transport of the surface floating oil (2DH
and 3D mode).
AF
- Density of water [kg/m3 ].
;
2
Wind parameters
; dd hh mm ss speed(m/s) direction(degr.)
0 12 30 0 6.000 270.000
0 20 30 0 6.000 270.000
- the number of “breakpoints” for the time function as an integer.
That number of records consisting of:
- 4 integers giving the time of a function point (day, hour, minute, second), followed by
DR
2 reals indicating the wind speed [m/s] and the direction (degrees from the North, e.g.
270 degrees is from the West).
At least two breakpoints should be given (also for zero wind velocities). The start time and
end time of the series of breakpoints (which must be a series increasing in time) should
be equal to the start time and stop time of the total simulation.
; Model specific parameters
; Oil model parameters

0 ;number of constants
- model specific constants (real).
For the tracer model, there are no additional constants and 0 should be entered. For the
oil module additional constants are required.
∆ That many records specifying the values of the constants (only required for the oil module,
see section 6.3.
; Timers
;yyyy mm dd hh mm ss
0 12 30 0 ;simulation start time
0 20 30 0 ;simulation stop time
0 0 0 0 ;DELWAQ take over time
0 12 30 0 ;map file start time
0 20 30 0 ;map file stop time
0 1 0 0 ;map file time step
0 12 30 0 ;his file start time
0 20 30 0 ;his file stop time
0 1 0 0 ;his file time step
1990 1 1 0 0 0 ;reference date for output
- Simulation timers (days, hours, min, sec).
Two sets of 4 integers, expressing the day, hour, minute and second of the start and the
stop of the simulation. The start time must be the same as the start time of the coupled
hydrodynamic output file. This is tested by the system. The end time may exceed the end
124 of 137 Deltares

time of the coupled hydrodynamic simulation. In that case, the hydrodynamics database
is rewinded automatically and re-used.
- D-Water Quality delay time (days, hours, min, sec).
One set of 4 integers defining day, hour, minute and second. This option is no longer
supported in the GUI and it is required to set the 4 integers to 0, otherwise unexpected
results may occur.
- Map file timers (days, hours, min, sec).
The map file is used for movie generation (as with the D-Water Quality map file). Three
sets of 4 integers must be provided, giving the day, hour, minute and second of the start of
the map file, the stop of the map file and the time step in the map file. This output gener-
T
ates concentration distributions mapped on the hydrodynamic grid (optionally aggregated
in the vertical).
- Time history file timers (days, hours, min, sec).
The time history file is used for time histories that can be plotted with GPP. Three sets
of 4 integers must be provided, giving the day, hour, minute and second of the start of
AF the file, the stop of the file and the time step in the history file. In this file a time-series
of concentrations is given at specified observation points (’monitoring stations’), specified
below.
- Time offset (year, month, day, hour, min, sec).
To map the model time to real world timings, 6 integers must be provided for the year,
month, day, hour, minute and second to be added to the model time to obtain the real
world time.
; Observation points
1
; x-coordinate y-coordinate
DR
'waste point' 13210.00 14000.00

- An integer giving the number of observation points, also called ’monitoring stations’.
The observation points are used to generate time histories that can be plotted in GPP (see
also the block ‘timers’ discussed above).
∆ For each observation point supply one line with three items, namely:
- a string of at most 20 characters giving the name of the observation point.
∗ the x-co-ordinate and y -co-ordinate of the observation point. The co-ordinates must
be consistent with the co-ordinate system of the hydrodynamic simulation.
NOTE: the z -co-ordinate of observation points need not be supplied since in 3D sim-
ulations always concentrations for all layers will be given in the history file.
; Output parameters
;Zoom grid output method
8
; dd hh mm ss recovery(-) (for scaling of output)
0 12 30 0 1.000
0 12 45 0 1.000
0 13 0 0 1.000
0 14 0 0 1.000
0 16 0 0 1.000
0 18 0 0 1.000
0 20 0 0 1.000
0 20 30 0 1.000
- Number of times to generate output using the zoom grid for concentration generation. The
principle of using the zoom grid is explained in Figure 4.12.2. What essentially is done is
to add the mass of all the particles in one grid cell of the zoom window and to calculate
the concentration by dividing this mass by the volume of the cell. This is done for all cells
in the zoom window (all layers).
∆ For every time, for which output using the zoom grid window is generated, 4 integers and
1 real must be provided. The integers specify the output times (in day, hour, minute, sec-
Deltares 125 of 137

ond). The real gives the recovery factor. A recovery factor of 1.0 yields concentrations as
modelled. A different recovery factor may be used to match the model results to measured
data, if not all material is recovered with the measurements.
; Zoom grid output parameters
5000.00 20000.00 ; x-start(m) x-end(m)
10000.00 25000.00 ; y-start(m) y-end(m)
150 150 ; grid cells in x dir(-) grid cells in y dir(-)
- x-co-ordinates of the zoom window.
- y -co-ordinates of the zoom window.
These values define the location and size of the zoom window. Output using the zoom
T
grid output method will be limited to this window and if particle stray outside this window
they will not be shown, even though these particles remain active in the model. The co-
ordinates are all expressed in the same units as used in the hydrodynamic computation
(for example national grid co-ordinates).
- Two numbers (integers) defining the number of grid cells in the x- and y -direction respec-
AF
tively.
;
1
Instantaneous releases
; number of instantaneous releases
'dye release 1'

0 12 30 0 ; release time (dd hh mm ss)
13210.00 17300.00 1 ; x-coord(m) y-coord(m) layer(-)
0 0.00 ; option for release radius radius[m]
; (0=user-defined radius;1=formula Fay-Hoult)
50.00 ; perc. of total particles(%)
;released mass(kg)
1.000e+008 ; Tracer
DR
0.000 ; Tracer con 52

- Number of instantaneous releases.
∆ that many sets as instantaneous releases consisting of:
- a string containing the name of the release (at most 20 characters).
- 4 integers for the release time (day, hour, minute, second).
- 2 reals for the x- and y -co-ordinates of the release point (use the same co-ordinate
system as the hydrodynamic grid).
∗ (on the same line) one integer giving:
a for K > 0: the layer of the release. For dye releases the particles are uniformly
distributed over the layer;
b for K = 0: a uniform release over depth.
- 1 integer to indicate if the radius is user-defined of calculated. Calculated is only
possible for the oil model).
∆∗ (on the same line) 1 real for the radius (in metres) of the release in the horizontal
x, y -plane. Particles are uniformly distributed within this circle.
- 1 real specifying the percentage of the total number of particles that is going to be used
to model the effects of this release. For instance, if you specified 10000 particles and a
percentage of 50.0 for this release, than 5000 particles will be released at this (x, y)-
co-ordinate (the mass corresponding to these particles must be specified below). The
total percentages of all releases (inclusive the continuous releases) may be smaller
or greater than 100 %, but in that case the number of particles released during the
simulation will be less or more than the specified number of particles.
- for each substance the total mass that is going to be released (each values on a new
line).
; Continuous releases
1 ; number of continuous releases
126 of 137 Deltares

'continuous release 1'

13210.00 14000.00 3 ; x-coord(m) y-coord(m) layer
0.000 50.00 ; radius(m) perc. of total particles(%)
1.000 ; scale factor for load (kg/s)
1 ; interpolation option (0=block 1=linear)
; released_concentrations (kg/m3)
0.000 ; Tracer
1.000e+005 ; Tracer con 52
; release rates
T
2
; dd hh mm ss rate (m3/s)
0 12 30 0 1.000
0 20 30 0 1.000
- One integer giving the number of continuous releases.
∆ Define that many sets as continuous releases:
AF- a string containing the name of the release (at most 20 characters)
- 2 reals for the x- and y -co-ordinate of the release point.
∗ (on the same line) one integer K giving, specified in the same way as for instanta-
neous releases.
- 1 real for the radius (in metres) of the release in the horizontal x, y -plane. Particles
are uniformly distributed within this circle.
∗ 1 real specifying the percentage of the total number of particles that is going to be
used to model the effects of this release.
- 1 real for the required scale factor on a new line.
- 1 integer to specify the interpolation option on a new line. A block function is specified
DR
by 0 and a linear interpolation by 1.

- For each continuous release specify the concentrations of all tracers. The concentra-
tions of the different tracers for a particular release are specified on a new line. The
information for the different locations is on separate lines. The order of the locations is
the same as the definitions of the releases.
- Now the mass to be released must be specified. This is done by means of specifying
breakpoints that are either interpolated using a block function (0) or using a linear interpo-
lation (1) — see the information on the interpolation option above.
- An integer specifying the number of breakpoints.
- For each breakpoint specify the time (days, hours, minutes, seconds) and the release
rate. For each breakpoint, the time and release rate of all releases are specified on
the same line.
; User defined releases
0
- Instantaneous release or User-defined releases.
This option is no longer supported and should be set to 0.
; Decay rates
2
;dd hh mm ss decay rates (1/day)
0 12 30 0 0.000 0.000
0 20 30 0 0.000 0.000
- one integer representing the number of "breakpoints" to specify the time variable decay
coefficient. At least two breakpoints are required. Each record consists of the following:
∆ 4 integers giving the time of a breakpoint (day, hour, minute, second).
∆ as many reals (on the same line as the breakpoint specification) as there are substances,
giving the first order decay coefficient per substance. The decay coefficient is assumed to
be specified in [d−1 ]. This information must be specified for at least two breakpoints. The
Deltares 127 of 137

first breakpoint must be equal to the start time of the calculation, and the last breakpoint
must be equal to the end time of the calculation.
; Settling velocities
; Generalized settling formula
0.0000 ; exponent for c(-)
1 ; grid refinement factor for accurate settling(-)
; Settling tables
2 ; table dimension
; Tracer
T
; dd hh mm ss A0[m/s] A1[m/s] Period[h] Phase[h] Vmin[m/s] Vmax[m/s]
0 12 30 0 1.000e-004 0.000 0.000 0.000 -10.000 1000.000
0 20 30 0 1.000e-004 0.000 0.000 0.000 -10.000 1000.000
; Tracer con 52
; dd hh mm ss A0[m/s] A1[m/s] Period[h] Phase[h] Vmin[m/s] Vmax[m/s]
0 12 30 0 0.000 0.000 0.000 0.000 -10.000 1000.000
AF 0 20 30 0 0.000 0.000 0.000
The time-variable settling velocities are specified in m/s.
0.000 -10.000 1000.000
- One real specifying the exponent for the concentration dependent settling (explained in
section 4.10.2).
- The grid refinement factor (integer), used to calculate the concentrations needed to drive
the concentration dependent settling velocity. The grid refinement factor is a multiplier
of the number of grid cells in the x- and y -direction (see the box Output Definitions). If
particles stray outside the zoom window, then the settling velocity will be set to 0.
- one integer representing the number of "breakpoints" for the time function for the (first
order) decay coefficient.
For each substance a block of data follows, specifying as many records as breakpoints.
DR
The following information should be present on each record:

∆ 4 integers giving the time breakpoint (day, hour, minute, second).
∆ four reals that specify the settling velocity: A0 [m/s], A1 [m/s], the period T [hours] and
phase φ [hours], followed by the minimum and maximum allowable settling velocity. All
values mentioned here are specified on the same line as the time of the breakpoint. For
more information on the implementation of the time-variable settling velocities, see sec-
tion 4.10.2.
;Sediment/erosion parameters
5.0000e-002 ; Critical shear stress for sedimentation (Pa)
4.0000e-001 ; Critical shear stress for erosion (Pa)
5.5000e+001 ; Chezy coefficient (m1/2/s)
∆ If the sedimentation/erosion option is switched on, the following constants must be given:
- Critical shear stress for sedimentation [Pa].
- Critical shear stress for erosion [Pa].
- Chézy coefficient [m1/2 /s].
Details of the sedimentation/erosion process are in section 4.10.2.
128 of 137 Deltares

A.3 Oil module constants

In the previous section the Sub-datagroup Oil Model did not contain any parameters because
the tracer model does not require any. When setting up the oil module, a number of specific
oil module parameters are required. In the box below an example is given of a model with two
oil types, three discharges for which the discharge radius of two discharges are calculated
instead of using the supplied discharge radius as given in the instantaneous and continuous
discharges boxes. For details of the model specific parameters see section 4.10.4.
; Oil model parameters
T
22 ;number of constants
; Oil fraction dependent parameters
; Ekofisk
0.4 ; Evaporation fraction per day
1 ; Oil dispersion option [0/1]
; (0=fraction per day 1=Delvigne/Sweeny formula)
0 ; Sticky probability [0,1]
AF 1
0
1
; Volatile fraction [0,1]
; Emulsification parameter C1
; Maximum water content C2 [0,1]
1 ; Evap. fraction at which emulsification starts
850 ; Density
50 ; Kinematic viscosity
; Heavy Fuel Oil
0.05 ; Evaporation fraction per day
1 ; Oil dispersion option [0/1]
; (0=fraction per day 1=Delvigne/Sweeny formula)
0.5 ; Sticky probability [0,1]
0.94 ; Volatile fraction [0,1]
2e-006 ; Emulsification parameter C1
DR
0.7 ; Maximum water content C2 [0,1]

0.05 ; Evap. fraction at which emulsification starts
990 ; Density
1500 ; Kinematic viscosity
; Global oil parameters
5e-005 ; minimum thickness oil layer
0 ; deflection angle (due to Coriolis force) - only for 3D
- an integer defining the number of model specific constants. For the oil module, this number
will vary depending on a number of options that are selected within this input box.
Following the specification of the number of constants for this block, for each oil type a block
of 8 parameters is specified (see section 6.5):
- the evaporation rate (per day).
- the choice of the formulation to calculate the oil dispersion (entrainment of oil in water).
The options are:
0 the dispersion is specified as a probability per day. The effect is an exponential reduc-
tion of surface floating oil.
1 the Delvigne/Sweeney formulation for the oil dispersion (entrainment) is used, which
means that the parameter that is specified is the kinematic viscosity [cSt].
∗ the dispersed fraction per day (option 0) or the value of the kinematic viscosity of the
oil (option 1).
- The probability of a particle to stick to the bottom or land boundary.
- The volatile fraction, i.e. the fraction of the oil that can evaporate.
- emulsification parameter C1 .
- Maximum water content (emulsification parameter C2 ).
- Evaporated fraction at which emulsification starts.
- Density of oil.
Deltares 129 of 137

- Kinematic viscosity of the oil.
Subsequent to the previous block follow some global oil parameters
- A real specifying the minimum oil layer thickness

∆ An integer specifying the deflection angle (in degrees, clockwise). This option is only
applied in the 3D mode and for a 2DH model this parameter is not included in the input
file.
T
A new setting not supported by the current GUI is a switch for sticking of oil at dry flats. To
include it, you will have to changing the first line to version V3.73.00 or higher, and add the
follow parameter, without increasing the number of constants:
0 ; sticking at drying flats [0,1]
A.4
AF ; (0=will not stick at dry flats, 1=will stick at dry flats)
Initial condition file for the oil module

The initial condition file (ASCII format) specifies one or more polygons that define surface oil
patches. Each individual patch can be generated using QUICKIN, by creating a polygon file
for each patch, for the file format see section A.5. If the polygon is not closed, then PART will
close it by connecting the first and last co-ordinate of the polygon. More than one patch may
be created. A header will then need to be added to each patch to specify the type and amount
of oil for this patch and the number of particles associated with this patch. The type of oil will
need to be the same as one of the oil types in the input file. An example of a initial conditions
DR
file is shown below.
*
* Fraction: Ekofisk
* Mass: 1000000 kg
* Particles: 10000
L00001
13 2
1.8809943E+05 6.1560469E+05
1.8784460E+05 6.1639466E+05
...
1.8860074E+05 6.1533856E+05
1.8809943E+05 6.1560469E+05
*
* Fraction: Heavy Fuel Oil
* Mass: 1000000 kg
* Particles: 20000
L00002
22 2
1.9166405E+05 6.2069936E+05
1.9486139E+05 6.2244163E+05
...
1.9110883E+05 6.1937831E+05
1.9166405E+05 6.2069936E+05
The keywords Fraction, Mass and Particles are compulsory. The ∗ at the beginning
of the line is also required. These three keywords are then followed by a specification of the
polygon name. The next line specifies the number of rows and number of columns that define
the polygon, followed by the x, y data.
This <∗.ini> file can be visualised in the visualisation tool of the PART-GUI, by importing the
file as a land boundary file.
130 of 137 Deltares

In the input file the use of initial oil patches looks like:
; Initial condition (oil only)
1 ; 0 = no initial particles
; 1 = initial oil patches (from *.ini file)
'ekofisk_hfo.ini' ; Initial condition file (including polygones describing the oil patches)
- An integer indicating if initial oil patches from a file are used.

∆ If for the oil module initial oil patches are required, then the name of this file is specified.
T
A.5 Polygon file
File contents The co-ordinates of one or more polygons. Each polygon is written
in a single block of data
Filetype ASCII
File format Free formatted
AF Filename
Generated
<name.pol>
RGFGRID, QUICKIN, D-WAQ DIDO, etc
Record description:
The file may contain one or more polygons. For every polygon the file should contain a line
indicating the name of the polygon, followed by a line indicating the number of points making
up the polygon and the number of coordinates, i.e. 2, finally followed by the coordinate data.
Record Record description

DR
Preceding description records, starting with an asterisk (∗), and will

be ignored.
1 A non blank character string, starting in column one
2 Two integers representing the numbers of rows and number of

columns for this block of data
Two reals representing the x, y or λ, ϕ-coordinate
Restriction:
⋄ The first record and the last record in the block should be the same
Example:
*
* Deltares, Delft3D-DIDO Version 3.39.01.4423:4459, Sep 25 2008, 20:10:54
* 2008-09-25, 22:11:08
*
Observation Area 001
5 2
1.8768018E+05 6.1708738E+05
1.8996981E+05 6.1001035E+05
1.9746314E+05 6.1266423E+05
1.9480925E+05 6.1838830E+05
1.8768018E+05 6.1708738E+05
5 2
2.0011703E+05 6.1818015E+05
Deltares 131 of 137

1.9819166E+05 6.1063479E+05
2.0568498E+05 6.0870942E+05
2.0797461E+05 6.1599460E+05
2.0011703E+05 6.1818015E+05
5 2
1.9340425E+05 6.1396516E+05
2.0183425E+05 6.1365294E+05
1.9944054E+05 6.0558720E+05
1.9522555E+05 6.0595146E+05
1.9340425E+05 6.1396516E+05
T
AF
DR
132 of 137 Deltares

A.6 Initial condition file for the oil module

The initial condition file (ASCII format) specifies one or more polygons that define surface oil
patches. Each individual patch can be generated using QUICKIN, by creating a polygon file
for each patch, for the file format see section A.5. If the polygon is not closed, then PART will
close it by connecting the first and last co-ordinate of the polygon. More than one patch may
be created. A header will then need to be added to each patch to specify the type and amount
of oil for this patch and the number of particles associated with this patch. The type of oil will
need to be the same as one of the oil types in the input file. An example of a initial conditions
file is shown below.
T
*
* Fraction: Ekofisk
* Mass: 1000000 kg
* Particles: 10000
L00001
AF 13
1.8809943E+05
1.8784460E+05
...
2
6.1560469E+05
6.1639466E+05
1.8860074E+05 6.1533856E+05
1.8809943E+05 6.1560469E+05
*
* Fraction: Heavy Fuel Oil
* Mass: 1000000 kg
* Particles: 20000
L00002
22 2
1.9166405E+05 6.2069936E+05
DR
1.9486139E+05 6.2244163E+05
...
1.9110883E+05 6.1937831E+05
1.9166405E+05 6.2069936E+05
The keywords Fraction, Mass and Particles are compulsory. The ∗ at the beginning
of the line is also required. These three keywords are then followed by a specification of the
polygon name. The next line specifies the number of rows and number of columns that define
the polygon, followed by the x, y data.
This <∗.ini> file can be visualised in the visualisation tool of the PART-GUI, by importing the
file as a land boundary file.
In the input file the use of initial oil patches looks like:
; Initial condition (oil only)
1 ; 0 = no initial particles
; 1 = initial oil patches (from *.ini file)
'ekofisk_hfo.ini' ; Initial condition file (including polygones describing the oil patches)
- An integer indicating if initial oil patches from a file are used.

∆ If for the oil module initial oil patches are required, then the name of this file is specified.
Deltares 133 of 137

T
AF
DR
134 of 137 Deltares

DR
AF
T
T
AF
DR
Photo by: Mathilde Matthijsse, www.zeelandonderwater.nl
PO Box 177 +31 (0)88 335 81 88

2600 MH Delft sales@deltaressystems.nl
Rotterdamseweg 185 www.deltaressystems.nl
2629 HD Delft
The Netherlands
DR
AF
T

Delft3D-PART User Manual

Uploaded by

Copyright:

Available Formats

Delft3D-PART User Manual

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Delft3D-PART User Manual

Uploaded by

Copyright:

Available Formats

Water quality and aquatic ecology modelling suite

Simulation of mid-field water quality and oil spills, us-

ing particle tracking

Published and printed by:

For sales contact: For support contact:

Copyright © 2023 Deltares

1 A guide to this manual 1

3.4.2 Coupling GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

4 Graphical User Interface 23

4.7 Numerical parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

4.13.1 Executing the water quality mid-field module D-WAQ PART . . . . . . 54

6 Conceptual model 101

6.6 Data requirements for the conceptual model . . . . . . . . . . . . . . . . . 114

7 Algorithmic implementation 117

A Input files D-WAQ PART 121

5.1 Specifications of the tutorial cases . . . . . . . . . . . . . . . . . . . . . . 55

3.1 Main window of Delft3D-MENU . . . . . . . . . . . . . . . . . . . . . . . . 11

4.1 Data Group Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

4.2 Data Group Hydrodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . 24

5.1 Coupling GUI - Data Group Description . . . . . . . . . . . . . . . . . . . 57

the Zoom grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

1.2 Manual version and revisions

1.3 Typographical conventions

Module Title of a window or a sub-window are in given in bold.

Save Item from a menu, title of a push button or the name of

<\tutorial\wave\swan-curvi> Directory names, filenames, and path names are ex-

delft3d-menu Commands to be typed by you are given in the font

In this User manual, user actions are indicated with this

1.4 Changes with respect to previous versions

2.15 Update of Friesian Tidal Inlet tutorial model<∗.mdp> file

2.14 Introduction of the <∗.mdp> file

– continued from previous page

2.12 MENU screens updated.

⋄ Emulsification is implemented in the model.

2.11 section 4.12.2: The use of a Zoom grid is compulsory.

2.00 Improved layout of manual.

2.1 Short description of the program functionality

In D-WAQ PART, two modules are available:

⋄ Tracer module: simulation of conservative or first order decaying substances.

⋄ the dynamics of patches close to an outfall location;

How does particle modelling work?

2.2 Specification of the program package

2.3 Installing the program

2.4 Error checking

2.5 Particle tracking module within Delft3D

Hydrodynamics Water Quality Particle Tracking ...

coupling PART-GUI Delft3D-PART QUICKPLOT

Figure 2.2: Software architecture around the particle tracking module

2.6 Input data files

2.7 Limitations in PART-GUI

Table 2.1: Limitations of the PART-GUI

Number Description Value

1 Maximum number of substances (oil counts as 3 substances) 90

2 Maximum number of continuous releases (note: number of instanta- 25

3 Maximum number of columns in (any) GUI table 70

4 Maximum number of oil types (oil counts as “1”) 30

5 Maximum number of time breakpoints in GUI table 1 000

6 Maximum number of layers in case of PLO output 100

3.1 Starting Delft3D