MIKE 21 Flow Model FM

ECO Lab / Oil Spill Module

User Guide

Powering Water Decisions MIKE 2023

2
PLEASE NOTE

COPYRIGHT This document refers to proprietary computer software which is pro-

tected by copyright. All rights are reserved. Copying or other repro-
duction of this manual or the related programs is prohibited without
prior written consent of DHI A/S (hereinafter referred to as “DHI”).
For details please refer to your 'DHI Software Licence Agreement'.

LIMITED LIABILITY The liability of DHI is limited as specified in your DHI Software
Licence Agreement:

In no event shall DHI or its representatives (agents and suppliers)

be liable for any damages whatsoever including, without limitation,
special, indirect, incidental or consequential damages or damages
for loss of business profits or savings, business interruption, loss of
business information or other pecuniary loss arising in connection
with the Agreement, e.g. out of Licensee's use of or the inability to
use the Software, even if DHI has been advised of the possibility of
such damages.

This limitation shall apply to claims of personal injury to the extent

permitted by law. Some jurisdictions do not allow the exclusion or
limitation of liability for consequential, special, indirect, incidental
damages and, accordingly, some portions of these limitations may
not apply.

Notwithstanding the above, DHI's total liability (whether in contract,

tort, including negligence, or otherwise) under or in connection with
the Agreement shall in aggregate during the term not exceed the
lesser of EUR 10.000 or the fees paid by Licensee under the Agree-
ment during the 12 months' period previous to the event giving rise
to a claim.

Licensee acknowledge that the liability limitations and exclusions

set out in the Agreement reflect the allocation of risk negotiated and
agreed by the parties and that DHI would not enter into the Agree-
ment without these limitations and exclusions on its liability. These
limitations and exclusions will apply notwithstanding any failure of
essential purpose of any limited remedy.

Powering Water Decisions 3

4 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S
CONTENTS

1 About This Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

1.1 Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.2 Assumed User Background . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3 General Editor Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3.1 Navigation tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.2 Editor window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.3 Validation window . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.4 Online Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3 Getting Started . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2 Project Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.3 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.4 New MIKE ECO Lab Template? . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.5 Hypothesis / Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.6 Implementation of Theory in the MIKE ECO Lab Template . . . . . . . . . . . 22
3.7 Set Up and Run Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.8 Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.9 Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.1 Oil Refinery Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.1.1 Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.1.2 Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
4.1.3 Parameters HD simulation . . . . . . . . . . . . . . . . . . . . . . . 29
4.1.4 Parameters - OS simulation . . . . . . . . . . . . . . . . . . . . . . 30
4.1.5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.1.6 Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.2 Extended Oil spill template, surface spreading . . . . . . . . . . . . . . . . . . 39
4.2.1 Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.2.2 Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.2.3 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.2.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.2.5 Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.3 Extended Oil spill template, boom example . . . . . . . . . . . . . . . . . . . 48
4.3.1 Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.3.2 Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

Powering Water Decisions 5

 4.3.3 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.3.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.3.5 Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
5 MIKE ECO LAB / OIL SPILL MODULE . . . . . . . . . . . . . . . . . . . . . . 61
5.1 Model Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.1.1 Template selection . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.1.2 MIKE ECO Lab Parameter set library . . . . . . . . . . . . . . . . . 62
5.1.3 Integration method . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.1.4 Update frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
5.1.5 Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
5.2 Classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
5.2.1 General description . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.2.2 Class specification . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.3 State Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.3.1 General description . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.3.2 Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.3.3 Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.4 Solution Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
5.4.1 Remarks and hints . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
5.5 Constants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.5.1 General description . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.5.2 Built-in constants . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.5.3 Remarks and hints . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.6 Forcings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.6.1 General description . . . . . . . . . . . . . . . . . . . . . . . . . . 72
5.6.2 Built-in forcings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
5.6.3 Remarks and hints . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.7 Dispersion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.7.1 Dispersion specification . . . . . . . . . . . . . . . . . . . . . . . . 73
5.7.2 Recommended values . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.8 Drift profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.8.1 General description . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.8.2 Class drift profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.9 Bed Roughness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.9.1 General description . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.9.2 Bed roughness from hydrodynamic model . . . . . . . . . . . . . . . 79
5.9.3 User specified bed roughness . . . . . . . . . . . . . . . . . . . . . 79
5.10 Wind Forcing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.10.1 User specified wind . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.11 Precipitation-Evaporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.11.1 Specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.11.2 Recommended values . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.12 Infiltration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.13 Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.13.1 Source specification . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.13.2 Remarks and hints . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

6 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

 5.14 Particle Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.14.1 Particle source specification . . . . . . . . . . . . . . . . . . . . . . 84
5.15 Initial Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
5.16 Boundary Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
5.16.1 Boundary specification . . . . . . . . . . . . . . . . . . . . . . . . . 90
5.17 Hydrodynamic Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.18 Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.18.1 User specified temperature . . . . . . . . . . . . . . . . . . . . . . . 92
5.19 Salinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.19.1 User specified salinity . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.20 Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5.20.1 User specified density . . . . . . . . . . . . . . . . . . . . . . . . . 93
5.21 Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5.21.1 Geographic view . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5.21.2 Output Specification . . . . . . . . . . . . . . . . . . . . . . . . . . 94
5.21.3 Output Items . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6 LIST OF REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

Powering Water Decisions 7

8 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S
Purpose

1 About This Guide

1.1 Purpose
The main purpose of this User Guide is to enable you to use, MIKE 21 Flow
Model FM ECO Lab Module, for applications involving ecological modelling.

1.2 Assumed User Background

Although the MIKE ECO Lab Module has been designed carefully with
emphasis on a logical and user-friendly interface, and although the User
Guide and Online Help contains modelling procedures and reference mate-
rial, common sense is always needed in any practical application.

In this case, "common sense" means a background in ecological modelling,

which is sufficient for you to be able to check whether the results are reason-
able or not. Proper use of this module requires knowledge about biological,
and chemical processes in the water environment at a level of environmen-
tally oriented engineers (civil engineer or chemical engineer) or model ori-
ented biologists. A co-operation between engineers and biologists is
recommended. It is assumed that you are familiar with the Hydrodynamic and
Transport modules of MIKE 21 Flow Model FM. This Guide is not intended as
a substitute for basic knowledge of the area in which you are working: envi-
ronmental modelling of aquatic systems.

Advanced users might want to create their own MIKE ECO Lab templates.
See the User Guide for MIKE ECO Lab available from the MIKE Zero Gen-
eral Documentation Index:

MIKE ECO Lab, User Guide

It is assumed that you are familiar with the basic elements of MIKE Zero: file
types and file editors, the Plot Composer, the MIKE Zero Toolbox, the Data
Viewer and the Mesh Generator. An introduction to these are available from
the MIKE Zero General Documentation Index:

MIKE Zero Pre- and Postprocessing, Generic Editors and Viewers,

User Guide

Plot Composer, User Guide

MIKE Zero Toolbox, User Guide

1.3 General Editor Layout

The MIKE Zero setup editor consists of three separate panes.

Powering Water Decisions 9

 About This Guide

1.3.1 Navigation tree

To the left there is a navigation tree showing the structure of the model setup
file, and it is used to navigate through the separate sections of the file. By
selecting an item in this tree, the corresponding editor is shown in the central
pane of the setup editor.

1.3.2 Editor window

The editor for the selected section is shown in the central pane. The content
of this editor is specific for the selected section, and might contain several
property pages.

For sections containing spatial data - e.g. sources, boundaries and output - a
geographic view showing the location of the relevant items will be available.
The current navigation mode is selected in the bottom of this view, it can be
zoomed in, zoomed out or recentered. A context menu is available from
which the user can select to show the bathymetry or the mesh and to show
the legend. From this context menu it is also possible to navigate to the previ-
ous and next zoom extent and to zoom to full extent. If the context menu is
opened on an item - e.g. a source - it is also possible to jump to this items edi-
tor.

Further options may be available in the context menu depending on the sec-
tion being edited.

1.3.3 Validation window

The bottom pane of the editor shows possible validation errors, and it is
dynamically updated to reflect the current status of the setup specifications.

By double-clicking on an error in this window, the editor in which this error

occurs will be selected.

1.4 Online Help

The Online Help can be activated in several ways, depending on the user's
requirement:

 F1-key seeking help on a specific activated dialog:

To access the help associated with a specific dialog page, press the
F1-key on the keyboard after opening the editor and activating the spe-
cific property page.

 Open the Online Help system for browsing manually after a specific
help page:

10 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Online Help

Open the Online Help system by selecting “Help Topics” in the main
menu bar.

Powering Water Decisions 11

 About This Guide

12 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

2 Introduction
MIKE ECO Lab is a numerical lab for ecological modelling. It is an open and
generic tool for customizing aquatic ecosystem models to describe water
quality, eutrophication, heavy metals and ecology. The module is mostly used
for modelling water quality as part of an Environmental Impact Assessment
(EIA) of different human activities. The tool is also applied in aquaculture for
e.g optimization the production of fish, seagrasses and mussels. Another use
is in online forecasts of water quality, see e.g.

http://www.waterforecast.com

The need for tailormade ecosystem descriptions is large, because ecosys-

tems vary. The strength of this tool is the easy modification and imple-menta-
tion of mathematical descriptions of ecosystems into the hydrodynamic mod-
els included in DHI Software.

The user can use predefined DHI supported MIKE ECO Lab templates, or
can choose to develop his own model concepts. The module can describe
dissolved substances, particular matter of dead or living material, living bio-
logical organisms and other components (all referred to as state variables in
this context).

The module is developed to describe chemical, biological and ecological pro-

cesses together with interactions between state variables and also the physi-
cal process of sedimentation of components can be described. State
variables included in MIKE ECO Lab can either be transported by advection-
dispersion processes based on hydrodynamics, or have a more fixed nature
(e.g. rooted vegetation).

Powering Water Decisions 13

 Introduction

14 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

General

3 Getting Started

3.1 General
The hydrodynamic basis for the MIKE ECO Lab Module must be calculated
using the Hydrodymamic Module of the MIKE 21 Flow Model FM modelling
system.

If you are not familiar with setting up a hydrodynamic model you should refer
to User Guide for the Hydrodynamic Module and the comprehensive step-by-
step training guide covering the Hydrodynamic Module of MIKE 21 Flow
Model FM. The user guide and the training guide (PDF-format) can be
accessed from the MIKE 21 Documentation index:

MIKE 21 Flow Model, Hydrodynamic Module, User Guide

MIKE 21 & MIKE 3 Flow Model FM, Hydrodynamic Module,

Step-by-Step Training Guide

A comprehensive training guide covering the MIKE ECO Lab Module of the
MIKE 3 Flow Model FM modelling system is also provided with the DHI Soft-
ware installation. The objective of this training guide is to show how to set up
an advanced MIKE ECO Lab model. The training guide (PDF-format) can be
accessed from the MIKE 3 Documentation index:

MIKE 3 Flow Model FM ECO Lab Module, Step-by-Step Training Guide

The purpose of this chapter is to give you a general check list, which you can
use when working with environmental projects using the MIKE ECO Lab
Module in MIKE 21 Flow Model FM.

The work will normally consist of the tasks shown in the diagram in
Figure 3.1.

As the diagram illustrates there can be different ways to reach the solution of
your project.

Some tasks are mandatory, such as:

 Project analysis: defining and limiting the MIKE ECO Lab model
 Data collection
 Set up and run the model
 Calibration
 Solution: Running the production simulations
 Solution: Presenting the results

Powering Water Decisions 15

 Getting Started

Other tasks in your project are optional and depend on your choice: do you
plan to create your own MIKE ECO Lab template or just use a predefined
one? In some cases a predefined template covering your specific problem
does not exist, and then a new MIKE ECO Lab template has to be developed.
If you choose to develop your own MIKE ECO Lab template, your project will
include the following additional tasks:

 Development of Hypothesis/Theory: Litterature study and formulation of

equations
 Implementation of theory into a MIKE ECO Lab template using the MIKE
ECO Lab editor

16 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Project Analysis

Figure 3.1 Working with MIKE ECO Lab

You can see a list of available predefined MIKE ECO Lab templates in the
MIKE Zero General Documentation Index.

3.2 Project Analysis

This task involves analysis and evaluation of the ecosystem problem and
solution strategy. The considerations should be about:

Powering Water Decisions 17

 Getting Started

1. What ecosystem state variables to simulate and the possible causal rela-
tions for their state. Describe which questions the model should be able
to answer.
2. If it is possible at all to describe the ecosystem state variables with the
available ecosystem knowledge, and if it possible to get sufficient data to
simulate the state variables.

Figure 3.2 Illustration of the project analysis phase: Defining the problem and its
causes and finding a solution strategy

The first thing to do is to point out the ecosystem variables that should be
described with a MIKE ECO Lab model. In environmental projects these are
often the variables that are considered as either dangerous, or vulnerable in
the ecosystem. Also it is often seen that MIKE ECO Lab models should
describe variables that are described in the criteria of environmental legisla-
tion. In aqua culture the state variables are usually those that affect the pro-
duction.

The next thing to do is to point out the possible causal relations for the state
of the system. The objective for most MIKE ECO Lab models is to be able to
quantify these causal relations and their effect on the ecosystem. The best
models tell us something that we do not already know even though we know
the possible causal relation of a problem. For instance if an area has oxygen
depletion, and experts agree that it is caused by eutrophication. Often they
can not agree who is to blame for the anoxia. Are the nutrients imported from
other areas and countries across the boundaries or are they coming from
local effluents? A MIKE ECO Lab model describing eutrophication of the area
would be able to give us such answers.

A MIKE ECO Lab model that describe the relation between cause and prob-
lem is also ideal for making prognostic scenarios. What happens to the oxy-
gen concentrations in an area if we reduce the local nutrient loadings? A
MIKE ECO Lab model describing eutrophication of the area would also be
able to give us that answer.

Finally, the project analysis should involve development of a conceptual dia-

gram of the ecosystem in question. It should be discussed if it is possible to

18 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Data Collection

get adequate data and if enough ecosystem knowledge about the ecosystem
exists to describe it.

3.3 Data Collection

This task involves the collection of data, field measurements and laboratory
analysis and it may take a long time if you have to initiate a monitoring pro-
gram. Alternatively it may be carried out very quickly if you are able to use
existing data which are immediately available.

Figure 3.3 The phase of collecting data, e.g. field measurements and laboratory
analysis

In all cases the following data should be collected:

 Bathymetric data such as charts from local surveys or, for example, from
the Hydrographic Office, UK, MIKE C-MAP, etc.
 Hydrodynamic boundary data: Salinity, temperature, water levels and /or
current velocities, which might be measurements or taken as model out-
put from larger regional models
 Hydrodynamic forcings, for example wind, air temperature and clearness
coefficient
 Hydrodynamic sources: Discharges from rivers and point sources
 MIKE ECO Lab boundary data: concentrations at the boundary of the
advective state variables in the MIKE ECO Lab template, which might be
measurements or taken as model output from larger regional models.
Note that the frequency of measurements at the boundary can be very
critical for the accuracy of the model
 MIKE ECO Lab forcings (depends on the content of the MIKE ECO Lab
template), for example solar radiation

Powering Water Decisions 19

 Getting Started

 MIKE ECO Lab loadings: Concentrations in sources and precipitation of

the advective state variables in the MIKE ECO Lab template. The quality
of these data is often very critical for the accuracy of the MIKE ECO Lab
model. Make sure you have all the point sources and a proper assess-
ment of the diffuse loading. Some times the loadings themselves are
model output from e.g. MOUSE or MIKE 11. There are also DHI tools for
estimating diffuse loadings that you might consider to use, if diffuse
urban loadings are important in your MIKE ECO Lab study.
 MIKE ECO Lab initial conditions of MIKE ECO Lab state variables. If the
retention time in the system is high, the initial conditions are quite impor-
tant, and should be measured. If the retention time is low the initial condi-
tions are not so important for the advective state variables, because the
system quite fast will “flush” the system anyway, and the concentrations
will be in balance after a short “warmup” period. Also the ‘not-moveable’
state variable (fixed) are often quite important to measure. Could for
example be maps of benthic vegetation, benthic fauna etc.
 Calibration data; these might be measurements from different monitoring
stations. They are used for comparing simulated values with measured
values. So these data are important for documenting the capability and
quality of the model.

3.4 New MIKE ECO Lab Template?

Is any predefined MIKE ECO Lab Template sufficient, or does the ecosystem
that you want to describe require a dedicated MIKE ECO Lab template?

DHI has a number of predefined MIKE ECO Lab templates.

 Eutrophication MIKE ECO Lab templates

These are predefined DHI supported MIKE ECO Lab templates that
include models of eutrophication processes in aquatic systems.

Eutrophication 1 is the classic DHI Eutrophication model. Eutrophica-

tion 1 can be used with an extended description of sediment and ben-
thic vegetation. Eutrophication 2 is a template that includes features
such as diurnal DO variations.

You have access to the scientific description of the different eutrophica-

tion templates from the MIKE Zero General Documentation index. It is
also possible to view the pdf file directly from the MIKE ECO Lab tem-
plate editor when the EU template is loaded.

20 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

New MIKE ECO Lab Template?

Figure 3.4 DHI has a number of predefined MIKE ECO Lab templates describing
different types of ecosystems

 Heavy Metal MIKE ECO Lab template

This is a predefined DHI supported MIKE ECO Lab template including

a model for estimating the spreading of heavy metals in aquatic sys-
tems.

You have access to the scientific description from the MIKE Zero Gen-
eral Documentation index. It is also possible to view the pdf file directly
from the MIKE ECO Lab template editor when the ME template is
loaded.

 Water Quality MIKE ECO Lab templates

These are predefined DHI supported MIKE ECO Lab templates that
include a series of simple water quality models.

State variables that are modelled with the WQ templates are:

– BOD
– DO
– Nitrate
– Nitrite
– Phosphate
– Faecal Coliform
– Total Coliform

You have access to the scientific description of the different water qual-
ity templates from the MIKE Zero General Documentation index. It is
also possible to view the pdf file directly from the MIKE ECO Lab tem-
plate editor when the WQ template is loaded.

Powering Water Decisions 21

 Getting Started

3.5 Hypothesis / Theory

If none of the existing predefined MIKE ECO Lab templates cover your prob-
lem, you should formulate your own theories for your own template.

Describe in mathematical terms the processes that affect the state variables
and how the state variables interact with each other.

Figure 3.5 The mathematical formulation of theory for a new MIKE ECO Lab tem-
plate usually requires a literature study

Ecosystem knowledge is usually collected by searching in literature.

3.6 Implementation of Theory in the MIKE ECO Lab Template

When you have formulated your mathematical description of an ecosystem,
you should implement it in a new MIKE ECO Lab template. You create a new
template with the MIKE ECO Lab editor and enter your expressions with the
editor.

22 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Set Up and Run Model

Figure 3.6 Implementation of ecosystem theory into a MIKE ECO Lab template is
easy with the MIKE ECO Lab editor

See the User Guide for MIKE ECO Lab available from the MIKE Zero Gen-
eral Documentation index.

3.7 Set Up and Run Model

Whether you created your own template or you picked one of the existing
templates, you are now ready to set up the MIKE ECO Lab model in MIKE 21
Flow Model FM.

Please see the step-by-step training guide for MIKE FM ECO Lab for learning
how to set up an advanced MIKE ECO Lab model. This can be accessed
from the MIKE 3 Documentation Index:

MIKE 3 Flow Model FM ECO Lab Module, Step-by-Step Training Guide

If you are not familiar with setting up a hydrodynamic model you should refer
to the Step-by-step training guide for the Hydrodynamic Module:

MIKE 21 & MIKE 3 Flow Model FM, Hydrodynamic Module,

Step-by-Step Training Guide

3.8 Calibration
By calibration of a MIKE ECO Lab set-up in any of the hydrodynamic models
included in DHI software it is attempted to find the best accordance between
computed and observed state variables (model components) by variation of a
number of parameters.

Powering Water Decisions 23

 Getting Started

Figure 3.7 Calibration phase: adjusting the model so that the model output fits the
measurements

Deviation between computed and observed state variables in MIKE ECO Lab
can be caused by a number of reasons. The most obvious are:

1. Inaccurate specification of concentrations at the model boundaries

2. Inaccurate specification of loadings to the ecosystem
3. Inaccurate initial concentrations
4. Inaccurate hydrodynamic set-up
5. Limitations in the chosen MIKE ECO Lab process descriptions
6. Inaccurate calibration of coefficients in MIKE ECO Lab process equa-
tions
7. Wrong measurements
8. Unstable numerical solution

The MIKE ECO Lab modeller should when calibrating a MIKE ECO Lab set-
up have the above-mentioned possible reasons for inaccuracy in mind,
before changing relevant parameters dramatically.

Calibrating coupled ordinary differential equations can be tricky, because of

the mutual dependency of the state variables. So if one state variable is cali-
brated to fit the observed values, maybe it will not fit so well after calibrating
other state variables in the ecosystem. Some dependencies are stronger
than others. Therefore it is advisable to calibrate the state variables with the
least dependencies to other state variables first.

If the output deviates from the measurements in a way that can not be
accepted the user has to identify if the problem is caused by a wrong ecosys-
tem theory, or if the reason is caused by another of the reasons mentioned
above. Whether the template has to be changed or not, this calibration task
will continue until the model describes the ecosystem in a causal way.

24 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Solution

3.9 Solution
When the calibration is satisfying, the last task remains: Running the produc-
tion simulations.

Now it is time to use the model and get answers from the model. Usually this
task will include simulation covering different scenarios.

MIKE Zero includes a number of advanced plotting facilities, that are useful in
this phase, when presenting your results. See the facilities in Plot Composer
and MIKE Animator Plus.

Powering Water Decisions 25

 Getting Started

26 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Oil Refinery Example

4 Examples
Please see the step-by-step training guide for MIKE 3 Flow Model FM ECO
Lab Module, for learning how to set up an advanced MIKE ECO Lab model
using MIKE 3 Flow Model FM ECO Lab Module. The approach is similar for a
2D model. The guide is available from the MIKE 3 Documentation index:

MIKE 3 Flow Model FM ECO Lab Module,

Step-by-step training guide

Further examples focusing on ABM modelling can be also be accessed from

the Documentation Index:

MIKE 3 Flow Model FM, ABM Lab - Drifter Example,

Step-by-step training guide

To demonstrate the capabilities of Oil Spill modelling some examples are

included with the installation of MIKE Zero. The directory names are as fol-
lows:

 Oil refinery (2D):

.\Examples\MIKE_21\FlowModel_FM\ELOS\Refinery_OS
 Extended Oil spill template, surface spreading (2D):
.\Examples\MIKE_21\FlowModel_FM\ELOS\Extended_OS
 Extended Oil spill template, Oil spill boom example (2D):
.\Examples\MIKE_21\FlowModel_FM\ELOS\Extended_OS

4.1 Oil Refinery Example

4.1.1 Purpose

Prøvestenen is an artificial island that was originally created for the location
of a fortress, whose mission was to protect the entrance to Copenhagen. In
1922 the area was converted to store oil and petrol. Today Prøvestenen offer
the Copenhagen oil port for the reception of tankers. Prøvestenen harbour
has a tank storage capacity of 1,000,000 m3 for inflammable liquids.

In january 2008 a significant oil spill from Prøvestenen occured where several
hundred thousand liters of Vacuum Gas oil ran out into the Sound. The acci-
dent happened when there was a hole in the hose to an oil tanker. The oil
flowed north past Hellerup, Skovshoved up to Ven. Luckily, due to the low
water temperature and the properties of the oil a large part of the oil could be
picked up in clumps by environmental ships during the first 24 hours after the
spill, but some of the oil drifted to settle on the coast between Helsingborg
and Landskrona in Sweden.

Powering Water Decisions 27

 Examples

With this accident in mind the Oil Spill model is setup to simulate the fate of
different types of oil from an accident and to investigate the influence of water
temperature on the weathering processes.The matrix of simulations is given
in Table 4.1

Table 4.1 Combinations of climatic conditions and oil type

Oil type Winter Summer

Heavy fuel SimA_1 SimA_2
Crude oil SimB_1 SimB_2
Gasoline SimC_1 SimC_2

4.1.2 Scenario

The hydrodynamic setup correspond to the setup described in the step-by-

step training guide covering the MIKE 21 Flow Model FM. This guide is avail-
able from the MIKE 21 Documentation Index:

MIKE 21 & MIKE 3 Flow Model FM, Hydrodynamic Module, Step-by-Step

Training Guide

The test conditions are:

 The sound area covers about 60 km x 100 km

 The simulation period for the oil spill is selected to 24 hours from mid-
night december 11th to midnight december 12th, during which the flow
direction shifts after 12 hours. At the same time the wind shifts from
south-westernly wind to northwesternly wind (see Figure 4.1).

28 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Oil Refinery Example

Figure 4.1 Descriptive conditions for oil spill simulation

Upper: Wind speed and direction
Lower: Simulated current speed and direction next to oil refinery

 The time step for the oil spill simulation is set to 300 s.
 The source spill event is simulated by a time-varying source positioned
at the location of the oil refinery: 200 m3 of oil was spilled during a 20
minutes period in the beginning of the simulation.
 Two water temperatures are investigated: winter conditions at 5 degrees
Celcius and summer conditions at 15 degrees Celcius.
 Three types of oil are investigated: Heavy fuel oil, Crude oil (Statfjord)
and Gasoline. The characteristics of each oil type are derived by the
information given in the ESTC Spill Technology Databases(1).

4.1.3 Parameters HD simulation

The main condition defining the hydrodynamic problem is:

 The model area is resolved by an unstructured mesh with 3636 elements

and 2057 nodes. The mesh is shown in Figure 4.8.
 An overall time step of 300 seconds is used and the duration of the HD
simulation is one day starting at 11/12/1993 00:00:00.
 Initial flow conditions are derived from a previous HD simulation.

1 http://www.etc-cte.ec.gc.ca/databases/OilProperties/oil_prop_e.html

Powering Water Decisions 29

 Examples

 The south and north boundaries are each defined with water levels spec-
ified as line series
 The wind forcing is included as varying in time but constant in domain.
The wind conditions are measured in Kastrup. The wind friction varies
with the wind speed using default values.
 The eddy viscosity has been chosen to Smagorinsky formulation with a
constant value of 0.24.
 The bed resistance type has been chosen as Manning number, with a
constant value of 45 m1/3/s
 Decoupled data files are output for future use.
 The resulting conditions (water level, total water depth and current condi-
tions) are saved to an output file.

Figure 4.2 Model area, the sound between Denmark and Sweden

4.1.4 Parameters - OS simulation

The main settings defining the oil spill problem are:

 The processes for the oil spill are described by the Oilspill_A3, DHI Oil
Weathering model, template.

30 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Oil Refinery Example

 The derived oil properties for the different oil types are shown in
Table 4.2 below..

Table 4.2 Oil properties for the three oil types in the simulations

Heavy Fuel oil Crude oil Gasoline

Distribution in (%):
Volatile oil 25 50 100
Heavy oil 47 40 0
Wax 2 8 0
Asphaltene 26 2 0

Distillation percentage at 1 26 95
180 deg.C
1st oil specific constant -0.12 2.67 13.2
2nd oil specific constant 0.013 0.06 0.21
Simple evaporation type 1 (quad) 0 (log) 0 (log)

Density (kg/m3)
Average 1010 831 753
Volatile 755 755 753
Heavy 1094 907 -
Dynamic Viscosity (cP) 13460 6 1
Viscosity coefficient tem- -0.136 -0.111 0
perature dependency

 Other processes are defined by default values.

 The dynamic viscosity values are given at reference temperature 15
degrees Celcius.
 The source spill of this class is simulated for an event located at (E:
351600 N: 6173500) The position of the source is shown in Figure 4.9.

Powering Water Decisions 31

 Examples

Figure 4.3 Location of Prøvestenen shown in detail of model area

 The 200 m3 of released oil corresponds to a total mass of each oil com-
ponent as given in Table 4.3...

Table 4.3 Total mass for each oil component in the simulations

Heavy Fuel oil Crude oil Gasoline

Distribution in (kg):
Volatile oil 37750 75500 150600
Heavy oil 102836 72560 0
Wax 4376 14512 0
Asphaltene 56888 3628 0

 60 particles are released per time step, corresponding to 25% of the oil
particles.
 The mass values are interpreted as flux. Hence the mass values in
Table 4.3 are divided by 1200 (20 x 60s) to obtain the flux values.
 The horizontal dispersion is included using a constant value in the dis-
persion coefficient formulation of 2.0.
 Vertical dispersion is not included
 The drift profile is defined by Surface Wind Acceleration with a wind
weight of 0.02 and a wind drift angle of 15 degrees.
 The wind forcing is defined as in the HD simulation
 The influence of waves is omitted from the calculation
 The first output file is a dfsu file containing the state variables. The dfsu
file is saved every 6 time step (30 minutes).
 The second output file is a Particle file (xml file) and will contain every 5th
particle saved every 6 time step.

32 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Oil Refinery Example

4.1.5 Results

The spill is released during the first 20 minutes of the simulation (4 time
steps). The currents combined with the drag force from the westerly wind at
the water surface may transport the oil away from the coast and out into
Oresund in the beginning of the simulation. When the wind and current
change direction the oil slick flows towards south. Figure 4.4 to Figure 4.6
shows the resulting particle tracks from the three simulations.

Figure 4.4 Particle tracks during simulation, Heavy Fuel oil type

Powering Water Decisions 33

 Examples

Figure 4.5 Particle tracks during simulation, Crude oil type

Figure 4.6 Particle tracks during simulation, Gasoline oil type

As seen from the three figures above the track is not the same for all the oil
types. Some of the oil particles reach the bottom during the 24 hour simula-
tion while others stay out into the water. And some particles evaporates

34 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Oil Refinery Example

before the end of the simulation. This is because of the different composition
of the oil types and the related difference in the weathering processes.The
horizontal dispersion adds to the spreading of the results.

The evaporation of volatile material typically takes place during the first hours
of an oil spill. To illustrate the process for this example, the Particle Track tool
in the Oil Spill Particle Track toolbox has been used for extracting specific val-
ues from the XML files to dfs0 files. Figure 4.7 shows the total mass of the
volatile components during the simulation for the three oil types, winter and
summer, respectively.

Figure 4.7 Mass of volatile components during simulation.

As seen from Figure 4.7 the volatile component of the oil quickly decrease
during the first part of the simulation for the lighter oils. The oil evaporates
more quickly in the summer than in the winter due to the higher temperature.
The volatile component in Gasoline evaporates entirely during the simulation,
whereas the evaporation rate for the volatile component of the Crude oil
seem to decrease slightly after the first hour. This behaviour can be explained
by weathering processes: as the oil density increase due to evaporation the
oil becomes more viscous and evaporation is slower. The settling velocity
changes and subsequently influence the movement of the particles. In the
case of heavy oil the evaporation is very slow but this also relates to the fact
that the heavy oil settles on the bottom very quickly, see Figure 4.8 and
Figure 4.9.

Powering Water Decisions 35

 Examples

Figure 4.8 Vertical speed of oil particles during simulation.

Figure 4.9 Vertical position of oil particle number 61 during simulation.

Figure 4.8 shows the vertical speed of the oil particles. This can be inter-
preted as a negative fall velocity. The lighter oil types remain buoyant during
the simulation even if the oil particles becomes less buoyant as the lighter
components evaporate. On the other hand the fall velocity for the heavy oil
particles decrease as emulsion processes cause uptake of water into the oil
particles.

As seen by Figure 4.9 the oil particles are released and the lighter particles
remain at the water surface whereas the heavy oil type sink quickly towards
the bed. When the oil particles are not at the water surface the evaporation

36 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Oil Refinery Example

process is hindered and hence the volatile component of the oil particle
remains constant, as seen in Figure 4.7.

By using decoupled files from a hydrodynamic simulation it is possible to

speed up an oil spill simulation as the hydrodynamic flow is already calcu-
lated. This can e.g. be used to quickly assess the impact of changing various
parameters.

The simulation of Heavy oil in winter is used as an example, where results for
particle numbers 11, 71, 131 and 191 has been extracted and shown below.
The oil spill specifications are changed such that it includes a vertical disper-
sion factor of 1, and the Drift profile is change to the type ‘Use bed shear pro-
file and surface wind acceleration’. The output is restricted to the first hour of
simulation, but with output each time step.

Figure 4.10 Heavy oil particle positions in domain for 4 specific particles

It can be seen from the figures that there are no larger impact from adding the
vertical dispersion in this case. The heavy oil particles generally reach the
bottom and settle within short time. Figure 4.11 and Figure 4.12 illustrates
that even if the oil particles has settled on the bed, the processes continues
and the water fraction of the oil particles will continue to increase until an
upper limit is reached.

Powering Water Decisions 37

 Examples

Figure 4.11 Vertical position of heavy oil particles during the first hour of simulation.

Figure 4.12 Content of water in heavy oil particles during the first hour of simulation

4.1.6 Files

The following data files (included in the \ELOS\Refinery_OS folder) are sup-
plied with MIKE 21 Flow Model ELOS FM:

File name: SimHD.m21fm

Description: MIKE 21 Flow Model FM specification file. Contains the HD
setup for generating decoupled files

File name: SimA1.m21fm, SimA2.m21fm

Description: MIKE 21 Flow Model FM specification files. Setups for Heavy oil,
winter conditions and summer conditions, respectively.

File name: SimB1.m21fm, SimB2.m21fm

Description: MIKE 21 Flow Model FM specification files. Setups for Crude oil
(type Statfjord), winter conditions and summer conditions, respectively.

File name: SimC1.m21fm, SimC2.m21fm

Description: MIKE 21 Flow Model FM specification files. Setups for Gasoline,
winter conditions and summer conditions, respectively.

38 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, surface spreading

File name: SimA1_Decoupled.m21fm

Description: MIKE 21 Flow Model FM specification file. Setup for Heavy oil,
winter conditions, using Flow conditions from decoupled files.

File name: ExtractParticleData.noilt

Description: Oil Spill Particle Track toolbox specification file

File name: OilSpill_A3.ecolab

Description: MIKE ECO Lab template for Oil Spill

File name: OilRelease.dfs0

Description: Contains state variables for oil source

File name: oresund.mesh

Description: Mesh file including mesh and bathymetry

File name: wind_kastrup.dfs0

Description: Wind speed and direction for area

File name: waterlevel_north.dfs1

Description: Water level variation at north boundary

File name: waterlevel_south.dfs1

Description: Water level variation at south boundary

File name: InitialFlowField.dfsu

Description: Initial flow field for HD simulation

4.2 Extended Oil spill template, surface spreading

4.2.1 Purpose

This example demonstrates the surface spreading of the oil particles in the
extended oil spill template. The extended template contains a mechanism
that ensures that oil particles are placed without overlapping if possible.
Please note that this spreading is independent on the surface spreading of
the oil itself (gravity spreading), i.e. simulated surface area of individual oil
particles will change due gravity spreading until a terminal thickness is
reached. Weathering process may further change a particles area due to
changes in the oil mass represented by a particle. The overall area covered
by all particles of the spill will be the result of the gravity spreading of the indi-
vidual particles and the demonstrated mechanism of spreading the particle
swarm to ensure no overlap of particles. An upper limit of the overall area
covered by all particles will be the sum of the individual particles surface area
if no overlap occurs.

Powering Water Decisions 39

 Examples

4.2.2 Scenario

The scenario is a fully artificial, hour glass shaped water body (two 2x2 km
basin connected by a narrow channel) of total 5x2 km and a constant water
depth of 10 m. An initial oil spill is released at the centre of the west basin in
the first time step. As there is no current movement, any large-scale move-
ment of oil particles is driven by wind drift. To allow the oil particles to spread
undisturbed wind first starts to blow from West in the second half of the simu-
lation period. Wind drift parameters are tuned to ensure a strait drift towards
East. Most oil weathering process are switched off as this simulation shall
only demonstrate the surface spreading mechanism.

4.2.3 Parameters

To demonstrate the surface spreading of oil particles most weathering pro-

cess are switched off by setting the relevant class constants of the "Oil" class
to zero as shown in Table 4.4 below:

Table 4.4 Weathering process parameters in Extended Oil Spill template

Name Description Unit Value

Enable_WaveDispersion Vertical wave induced disper- logic switch 0
sion: on-off switch (0=off, 1=on) (0/1)
Enable_EVAP Evaporation: on-off switch logic switch 0
(1=on, 0=off) (0/1)
Enable_BIOD Biodegradarion: on-off switch logic switch 0
(1=on, 0=off) (0/1)
Enable_PHOT Photo-oxidation: on-off switch logic switch 0
(1=on, 0=off) (0/1)
Enable_Diss_s Surface dissolution: on-off logic switch 0
switch (1=on, 0=off) (0/1)
Enable_Diss_wc Water column dissolution: on- logic switch 0
off switch (1=on, 0=off) (0/1)
Enable_Wateruptake Water uptake: on-off switch logic switch 0
(1=on, 0=off) (0/1)
Enable_Waterrelease Water release: on-off switch logic switch 0
(1=on, 0=off) (0/1)
Enable_Diameterchange Dropplet diameter change: on- logic switch 0
off switch (1=on, 0=off) (0/1)
Enable_Areachange Area change: on-off switch logic 1
(1=on, 0=off) switch (0/1)
Enable_Boom Boom handling: on-off switch logic switch 0
(1=on, 0=off) (0/1)

40 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, surface spreading

Table 4.4 Weathering process parameters in Extended Oil Spill template

Name Description Unit Value

Enable_Dispergent Dispergent handling: on-off logic switch 0
switch (1=on, 0=off) (0/1)
Enable_In_Situ_Burning In-Situ burning handling: on-off logic switch 0
switch (1=on, 0=off) (0/1)

The surface spreading is controlled by the parameters listed in Table 4.5:

Table 4.5 Factors controlling the surface spreading

Name Description Unit Value

ShuffleSpeedFactor Disc spreading speed factor undefined 300
rfak Factor for area search function undefined 3

The "ShuffeSpeedFactor" controls the speed overlapping oil particles move

apart. In principle two overlapping particles will move apart from each other
along the line connecting the centres of the particles. The overall speed is
proportional to the overlap area and modified by the "ShuffeSpeedFactor";
lower values will cause a faster movement whereas higher values a lower
speed. If unrealistic spreading patterns are observed, the "ShuffeSpeedFac-
tor" should be adjusted, i.e. if insufficient spreading is detected the value
should be lowered whereas with too much spreading(1) the value should be
increased. The value is sensitive for the overall simulation time step and
should be revisited when timing changes are done.

The "rfak" value influences which particles are tested for overlap. The surface
spreading tests the neighbouring particles which centres that are more than
"rfak * own_radius" away. The factor should only be modified if oil particles of
quite different size interact (fx. particles from different spill events).

4.2.4 Results

Visualising surface spreading

To visualise the spreading effect the particle track is shown as an overlay
over the dfsu output. The style settings shown in Figure 4.13 are applied to
the instantaneous particle overlay.

1 In extreme cases a too fast spreading will result in an oil "ring" around the geometric centre of the particles

Powering Water Decisions 41

 Examples

Figure 4.13 Style for visualising the particles

The oil particles are visualised as circles ("Marker Symbol") with a variable
size ("Marker size type"). The size is derived from the oil particle variable
"Diameter for equvalent circular slick(m)" with a scaling factor of 0.5 ("Marker
scale factor").

To visualise the spreading speed the variable "Check if own radius and near-
est particle overlap" is used to derive the particle colour. In real simulations
one would probably use another property like the oil mass or thickness etc.

Shortly after the spill the oil particles are staged over each other and occupy a
small area (see Figure 4.14). The dark blue colour of the particles represents
a high overlap, particles with no overlap stress will appear red.

42 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, surface spreading

Figure 4.14 Particles shortly after the spill, time step 20

Background contours illustrates concentration

Some time into the simulation the surface spreading has caused the oil parti-
cles to move apart (see Figure 4.15). The border particles show much less
overlap compared to the particles at the centre of the spill. Please note that
just a very gentle dispersion is used no other movement due to currents or
wind is active yet.

Figure 4.15 Particles some time into the simulation, time step 100

Powering Water Decisions 43

 Examples

Directly before the wind sets in the oil slick has spread so much that there is
very few overlap in the oil particles (see Figure 4.16).

Figure 4.16 Particles just prior to the wind begins to affect movement, time step 157

When the wind sets in the whole spill moves East (see Figure 4.17). Note that
the overlap in particles is even more reduced.

Figure 4.17 Particles when wind has set in, time step 289

44 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, surface spreading

When the wind pushed the oil spill into the channel some particles are
blocked and accumulate at the shore (see Figure 4.18). This causes a high
overlap detection.

Figure 4.18 Particles when channel has been reached, time step 317

Effect of the "ShuffeSpeedFactor"

The following figures represent the effect of the "ShuffeSpeedFactor" value.
In Figure 4.19 the factor is quite low and an artificial "ring" spreading pattern
can be observed in the 2nd step

Powering Water Decisions 45

 Examples

Figure 4.19 Low ShuffleSpeedFactor, particles in step 2

After 100 timesteps the particles are spread over a large area and show
some "unnatural" jumping (compared to the corresponding image in
Figure 4.15).

Figure 4.20 Low ShuffleSpeedFactor, particles in step 100

In the following figures the spreading factor was set to a high value. Conse-
quently, the particles are not spread much at time step 100, see Figure 4.21.

46 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, surface spreading

Figure 4.21 High ShuffleSpeedFactor, particles in step 100

Also, later in the simulation the particle swarm does not spread much and
reaches the channel with a large overlap in all particles (see Figure 4.22). It
almost passes the channel "en bloc"

Figure 4.22 High ShuffleSpeedFactor, particles in step 317

Powering Water Decisions 47

 Examples

4.2.5 Files

The following data files (included in the \ELOS\Extended_OS folder) are sup-
plied with MIKE 21 Flow Model ELOS FM:

Filename: hourglass.mesh
Description: Mesh file including the mesh and bathymetry

Filename: Oilspill_Extended.ecolab
Description: copy of extended oilspill template (ECO Lab Template)

Filename: SurfaceSpreadingExample.m21fm
Description: MIKE 21 Flow Model FM specification file

Filename: WindSpeed.dfs0
Description: Wind data used in simulation

4.3 Extended Oil spill template, boom example

4.3.1 Purpose

The extended template contains a mechanism that simulates the application

of containment and diversion oil booms. This example demonstrates how to
define and use a boom to block the surface drift of oil particles in the
extended oil spill template.

4.3.2 Scenario

The scenario is a full artificial, hour glass shaped water body (two 2x2 km
basin connected by a narrow channel) of total 5x2km and a constant water
depth of 10 m. An initial oil spill is released at the centre of the west basin in
the first time step. As there is no current movement, any large-scale move-
ment of oil particles is driven by wind drift. To allow the oil particles to spread
undisturbed wind first starts to blow from West in the second half of the simu-
lation period. Wind drift parameters are tuned to ensure a strait drift towards
East. Most oil weathering process are switched off. It is basically the same
scenario used in the surface spreading example described on page 39.

48 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, boom example

4.3.3 Parameters

To demonstrate the use of a boom definition to block the spreading of oil par-
ticles most weathering process are switched off by setting the relevant class
constants of the "Oil" class to zero as shown in Table 4.6 below:

Table 4.6 Weathering process parameters in Extended Oil Spill template

Name Description Unit Value

Enable_WaveDispersion Vertical wave induced disper- logic switch 0
sion: on-off switch (0=off, 1=on) (0/1)
Enable_EVAP Evaporation: on-off switch logic switch 0
(1=on, 0=off) (0/1)
Enable_BIOD Biodegradarion: on-off switch logic switch 0
(1=on, 0=off) (0/1)
Enable_PHOT Photo-oxidation: on-off switch logic switch 0
(1=on, 0=off) (0/1)
Enable_Diss_s Surface dissolution: on-off logic switch 0
switch (1=on, 0=off) (0/1)
Enable_Diss_wc Water column dissolution: on- logic switch 0
off switch (1=on, 0=off) (0/1)
Enable_Wateruptake Water uptake: on-off switch logic switch 0
(1=on, 0=off) (0/1)
Enable_Waterrelease Water release: on-off switch logic switch 0
(1=on, 0=off) (0/1)
Enable_Diameterchange Dropplet diameter change: on- logic switch 0
off switch (1=on, 0=off) (0/1)
Enable_Areachange Area change: on-off switch logic 1
(1=on, 0=off) switch (0/1)
Enable_Boom Boom handling: on-off switch logic 1
(1=on, 0=off) switch (0/1)
Enable_Dispergent Dispergent handling: on-off logic switch 0
switch (1=on, 0=off) (0/1)
Enable_In_Situ_Burning In-Situ burning handling: on-off logic switch 0
switch (1=on, 0=off) (0/1)

To simulate containment and diversion oil booms a special data file ("boom_-
file.csv") must be present in the simulation root folder. This text file contains
the definition of the boom nodes and line segments as well as the skimmer
pump capacity over time. This means that the boom position can be
adjusted/moved throughout the simulation.

The easiest way to produce a valid boom data file is to utilise the special "Oil-
spill Boom and Skimmer" tool in the Oil Spill Particle Track toolbox. This tool
allows to define the boom position and features using a graphical input editor.

Powering Water Decisions 49

 Examples

When run the tool interprets the user input and generates a proper boom data
file.

In general, the data file is a plain text file. All data is separated by a comma.
After a header line at least one or multiple data definition lines follow. A data
line describes the position of one or more boom segments at a time step.

The header line consists of 4 integer digits, ndata, max_seg, utm_zone, pro-
jection, as outlined in Table 4.7.

Table 4.7 Parameters in headline of ascii boom file

Variable Description
ndata Total number of data lines in the file
max_seg Max number of segments defined in any data line
utm_zone Local UTM zone used for coordinates if not using model zone
projection Flag if model coordinates are UTM or long/lat
(projection=1 'model is in UTM and local utm zone is the same
projection=0 'model is in long/lat and coordinates are specified in
local utm zone coordinates)

Each data line consists of a time stamp, the number of boom segments in the
line and one or more boom segment definition. No more than max_seg seg-
ment definitions may occur in a data line!

{Timestamp}, NSeg, {boom segments}

A timestamp consists of 6 integer digits: YYYY, MM, DD, HH, MM, SS

Table 4.8 Parameters in time stamp

Variable Description
YYYY Yeat in 4 digits
MM Month number (1=January, 12=December)
DD Day number in month (1-31)
HH Hour (0-23)
MM Minutes (059)
SS Seconds (0-59)

After the timestamp the number NSeg, i.e. the number of boom segment defi-
nitions in the data line follows. Each of these boom segments is defined by 7

50 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, boom example

numbers, AX, AY, BX, BY, Pump_ID, Pump_capacity, Lock_prob, as

described in Table 4.9 below.

Table 4.9 Parameters for boom segments

Variable Description
Ax X coordinate of the segment start node A
Ay Y coordinate of the segment start node A
Bx X coordinate of the segment start node B
By Y coordinate of the segment start node B
Pump_ID ID of pump associated with the current line segment
(1-n, integer number)
Pump_capacity Capacity (in kg) of the pump (float number)
Lock_prob Probability of any oil particle being locked/catched by the boom
(0-1.00 = 0-100%) in a time step (float number)

The use of individual segments allows the definition of spatial independent

booms.

A boom always floats on the surface and extends down into the water col-
umn. The boom thickness can be set in the template constant "BoomDepth"
and is 1 m by default. Oil particles passing the surface below the boom are
not affected.

The template constants outlined in Table 4.10 control the boom efficiency and
effects.

Table 4.10 Template constants affecting boom efficiency and effects

# Description Default
11 Maximum windspeed (m/s) for boom operation 0 m/s
12 Efficiency of skimmer pumps during daytime opera- 100 %
tion
13 Efficiency of skimmer pumps during nighttime opera- 80 %
tion
14 Wave height below no leakage from booms is 2 m
observed
15 Wave height where booms become ineffective due to 4 m
wave action
16 Minimum boom towing speed through water 0.05 m/s
17 Maximum boom towing speed through water 0.4 m/s
18 Viscosity where boom leakage due to low viscosity 0.5 cP
stops
19 High viscosity limit in booms before leakage 0.7 cP

Powering Water Decisions 51

 Examples

Table 4.10 Template constants affecting boom efficiency and effects

# Description Default
20 A reduction factor in boom effectiveness at high Vis- 0.2 Fraction
cosity
21 Minimal oil slick thickness before skimmer pumps 0.1 mm
become ineffective
22 Boom Depth 1 m

Take care that these constants are set properly (especially #11, #16, #17 and
#22), otherwise unexpected results (ineffective boom) may be observed.

Using the Oilspill Boom and Skimmer tool

To generate the boom definition file one can use the “Oilspill Boom and Skim-
mer" tool. If this tool is invoked the user has to provide a valid tool setup
name for the tool as shown in Figure 4.23.

Figure 4.23 Intiial setup dialog for Boom and Skimmer tool in Oil Spill / Particle
Tracking toolbox

On the following dialog the domain can be either read from a mesh file or be
imported from an existing setup.

52 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, boom example

Figure 4.24 Dialog for specifying domain area for computation

When importing from a setup the time step settings are also imported to
match the setup data. Otherwise the user has to provide a proper time step
setting.

Figure 4.25 Dialog for specifying time for simulation

Now the user can place one or several booms in time and space

Powering Water Decisions 53

 Examples

Figure 4.26 Dialog for specifying boom segments in the domain

To insert a new boom segment at the active time step, use the "New" button.
This will add a boom to the definition table and set it as active boom. The
table will just show the start and end point of a boom. To modify individual
points of the boom, use the "Edit" button. This will bring up an extra dialogue
to modify the individual segments of the boom. You can also use the right-
click context menu in the graphic representation to modify the active boom
(i.e. add/move/delete nodes, duplicate etc.). The context menu allows also to
add a particle track overlay for placing a boom in a oil spill response planning.
The time slider can be used to set the active time; note the display shows all
time steps and a time stamp is placed next to the start node of each defini-
tion.

54 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, boom example

When creating the boom definition file, the node positions will be interpolated
if there are more than one timestamps defined, i.e. one need just to add key
frame positions. Pay attention to the fact that the boom will be removed after
the last timestamp. If just one timestamp is defined the boom will persist
throughout the whole simulation.

When finished with the boom specification the output file has to be defined.

Figure 4.27 Dialog for specifying output file

It is best to place this file with the default name "boom_file.csv" in the simula-
tion root folder. Please note that file name and location used by the oil spill
simulation is mandatory. If the result file is named different or placed in a dif-
ferent location it must be manually copied and/or renamed before it is used.

The following status page (Figure 4.28) gives a short overview on the defined
booms and allows the user to generate the output file by pressing the "Exe-
cute" button. The same will happen if the tool definition is selected and "run"
is pressed in the overview.

Figure 4.28 ‘Run’ command or Execute from Status page in simulation setup editor

Powering Water Decisions 55

 Examples

Once the boom definition file has been generated the main simulation can be
run.

4.3.4 Results

There is currently no direct way to use the boom definition file to visualise the
results. If you want to visualise a boom you can make use of a XYZ-file or
dfs1 format.

These file types can then be used in the Data Manager or Result Viewer to
visualise a boom. For some technical reasons the Result Viewer is more
suited as it can connect the boom nodes. The XYZ format is suitable for sta-
tionary booms whereas the dfs1 format can be used for moving booms as
well.

To visualise the stationary example boom the node coordinates have to be

saved in a XYZ file. This is a plain text file with 4 whitespace separated val-
ues representing the x,y,z -coordinates and a connectivity number. An exam-
ple XYZ file with the coordinates of the boom this looks like:

2050.00 1230.00 0.00 1

2067.02 1006.9 0.00 1
2047.69 881.274 0.00 0

To be able to visualise the boom the connectivity (4th value) must be speci-
fied such that the value in the boom end point in the last line is zero and the
other lines have "1".

The file can be added to a Result Viewer view similar as a standard dfsu and
the particle file. To draw a line between the points you need to enable it on the
line style properties page of the XYZ data.

56 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, boom example

Figure 4.29 Specifying line display based on boom coordinates from xyz data

In the example the flag "Set to one to lock a particle at a boom" is used to col-
our the particles representing a particles state of being lock at a boom or
freely moving on the surface.

The screen dumps in the following figures shows basically the same data as
for the surface spreading example. However, a boom is present that block the
entrance to the channel connecting the two basins. The oil particles are
driven by westerly winds towards the barrier.

Figure 4.30 Particles before reaching boom

For the time step shown in Figure 4.30 no particles are blocked by the boom
yet (compare to the image shown in Figure 4.17). The following figures show

Powering Water Decisions 57

 Examples

the oil plume just few time steps later and at the end of the simulation. Here a
large number of particles are become blocked by the boom (indicated by the
red colour).

Figure 4.31 Particles when first reaching the boom

Figure 4.32 Particles at the end of the simulation

58 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Extended Oil spill template, boom example

Note: the XYZ file can just be used to visualise stationary booms. If booms
move in time and space the coordinates can be provided as dfs1 file and plot-
ted similar. Here you need to provide a dfs1 file containing at least 2 items of
the type "Geographical coordinate" representing the x/y coordinates of the
boom nodes. Add as many grid points as boom nodes. You can just provide
keyframe coordinates, i.e. there will be linear interpolation between coordi-
nates in time. When visualising the dfs1 in the Result Viewer select
"Type=Spatial line" and "Location=Geographical position" as well as the
proper X and Y items in the spatial line properties.

4.3.5 Files

The following data files (included in the \ELOS\Extended_OS folder) are sup-
plied with MIKE 21 Flow Model ELOS FM:

Filename: hourglass.mesh
Description: Mesh file including the mesh and bathymetry

Filename: Oilspill_Extended.ecolab
Description: copy of extended oilspill template (ECO Lab Template)

Filename: BoomExample.m21fm
Description: MIKE 21 Flow Model FM specification file

Filename: WindSpeed.dfs0
Description: Wind data used in simulation

Filename: BoomDefinition.noilt
Description: Oil Spill Particle Track toolbox file containing an oil spill boom
and skimmer definition to generate the boom data file

Filename: boom_file.csv
Description: Boom data file generated by the oil spill boom and skimmer defi-
nition

Filename: Boom.xyz
Description: manually created XYZ file containing the stationary boom nodes
for visualisation purpose only.

Filename: BoomLine.dfs1
Description: manually created dfs1 file containing a moving boom for visualis-
ation purpose only.

Filename: BoomVisualisation.rev
Description: Result viewer setup for the simulation results, showing the simu-
lation area, oil particles and the stationary boom using a XYZ-file.

Filename: BoomVisualisation_lines.rev
Description: Result viewer setup for the simulation results, showing the simu-
lation area, oil particles and a moving boom using a dfs1-file.

Powering Water Decisions 59

 Examples

60 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Model Definition

5 MIKE ECO LAB / OIL SPILL MODULE

MIKE ECO Lab is a numerical lab for ecological modelling. It is an open and
generic tool for customizing aquatic ecosystem models to describe water
quality, eutrophication, heavy metals and ecology using process oriented for-
mulations.

Besides the pure process orientated description MIKE ECO Lab also sup-
ports the characterisation of individual spatial entities (or particles) with asso-
ciated attributes, processes and description of their movement. Such entities
can represent different individuals like fish or particles like oil parcels.

5.1 Model Definition

In this dialog you specify a MIKE ECO Lab template, representing the actual
model formulation and the solution parameters: Integrated Method and
Update Frequency.

5.1.1 Template selection

A MIKE ECO Lab template contains mathematical equations describing an

aquatic ecosystem. The user has two options:

 Using one of the DHI supported predefined MIKE ECO Lab templates or
the DHI supported predefined oil spill templates
(Note: the predefined oil spill templates must be imported using ‘From
file...’, and can be found in the installation folder under .\MIKE
Zero\2023\Templates\OS)
 Using another MIKE ECO Lab template or oil spill template. Choose
‘From file...’

Having selected the template a brief summary of the contents is shown in the
dialog.

Please note that every time a new template is selected, the specifications of
all the remaining MIKE 21 ECO Lab / Oil Spill dialogs are reset to default val-
ues. Especially the user parameterisation of constants and forcing is reset to
the default values stored in the template. However, the MIKE ECO Lab
Parameter set library provides an easy mechanism to store and retrieve a
complete parameterisation and makes reloading/updating of a template sim-
ple.

Predefined MIKE ECO Lab templates

The installation contains a number of standard MIKE ECO Lab templates.
They can be used for a variety of process oriented investigations as well as
investigations concerning the behaviour of individual entities.

Powering Water Decisions 61

 MIKE ECO LAB / OIL SPILL MODULE

You can find a description of the available templates in the MIKE Zero Docu-
mentation Index that can be accessed from the start menu. This is also where
you can find a description of the available Oil Spill templates.

Load Parameter set

You can load a set of previous specified parameters, i.e. populate the MIKE
ECO Lab constants and forcing settings with previous specified values. You
can also invoke the MIKE ECO Lab Parameter set library by pressing the
library button. After applying a parameter set a short summary will be shown,
listing the successful matched parameters and occurred errors/warnings.
Please note that the matching algorithm operates on the MIKE ECO Lab
symbol names and element types, i.e. it is possible that a collection and the
setup contain different numbers of specified parameters or may be based on
different MIKE ECO Lab templates (versions). If parameters refer to data files
the user must play attention to make these data files available.

5.1.2 MIKE ECO Lab Parameter set library

The MIKE ECO Lab parameter set library can be used to maintain different
parameterisations, i.e. value settings for the MIKE ECO Lab constants and
forcing. A single parameterisation, i.e. the specific value settings for all the
(Eulerian and ABM) MIKE ECO Lab constants and forcing is called a parame-
ter set. Such sets can be stored (and retrieved) from individual files, so called
parameter set collections. The parameter set library consist of at least one or
more such collections.

The default collection is the startup collection and represents the parameteri-
sation of the initially loaded setup. It is neither possible to add nor to remove
this default collection or to add/delete sets to it. The main purpose of the
startup collection is be able to revert to the initial states stored in the setup (it
does not necessarily reflect the defaults/initial value settings of the template).
To add a new, user defined collection use the "new" button and specify the
storage location and file name in the following dialogue. Please note that
existing files will be overwritten without further notice! The "load" button can
be used to open an existing parameter collection file, i.e. make all stored set-
tings available in the load parameter list (see Template selection). 'The
"unload" button will remove the access to selected collection and all its sets
but will not delete the collection file.

To add the current active value specifications of the MIKE ECO Lab setup,
i.e. the constants and forcing specification, to a collection, select the collec-
tion and press the "add" button. This will append the collection and add a new
parameter set. You can specify a name and a short description text for both
the collection and for each stored parameter set that will be used to identify a
specific set. To update an existing parameter set with the current value set-
tings, select the set and use the "update" button. This will overwrite the stored
specifications with the current values from the setup.

62 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Model Definition

To leave the parameter set library use the "close" or "save and close" buttons.
The later will write any new/updated set (marked with a "*" in front of their
name) to the specified collection files.

5.1.3 Integration method

The specification of the solution parameters includes selection of the integra-

tion method for the coupled ordinary differential equations defined in the .eco-
lab file. At present the following three methods are available:

 Euler integration method

 Runge Kutta 4th order
 Runge Kutta 5th order with quality check (not available in ABM Lab)

Euler integration method

A very simple numerical solution method for solving ordinary differential
equations.

The formula for the Euler method is:

y n + 1 = y n + h  f  x n y n  (5.1)

which advances a solution y from xn to xn+1 = xn + h

Runge Kutta 4th order

A classical numerical solution method for solving ordinary differential equa-
tions. It has usually higher accuracy than the Euler method, but requires
longer simulation times. The fourth order Runge-Kutta method requires four
evaluations of the right hand side per time step.

y n + 1 = rk4  y n f  x n y n  x n h  (5.2)

Powering Water Decisions 63

 MIKE ECO LAB / OIL SPILL MODULE

The function is solved this way:

k 1 = h  f  x n y n  (5.3)

h k
k 2 = h  f  x n + --- y n + ----1-
 2 2

h k
k 3 = h  f  x n + --- y n + ----2-
 2 2

k 4 = h  f  x n + h y n + k 3 

k k k k
y n + 1 = y n + ----1- + ----2- + ----3- + ----4- – O  h 5 
6 3 3 6

which advances a solution y from xn to xn+1 = xn + h

Runge Kutta 5th order with quality check

A numerical solution method for solving ordinary differential equations. The
accuracy is evaluated and the time step is adjusted if results are not accurate
enough.

y n + 1 = f  y n f  y n x n  x n h  yscale  (5.4)

The function is solved this way:

First take two half steps:

h 2 = 0.5  h (5.5)

xn + ½ = xn + h2 (5.6)

y 2 = rk4  y n f  y n x n  x n h 2  (5.7)

y 2 = rk4  y 2 f  y 2 x n + ½  x n + ½ h 2  (5.8)

Compare with one full time step:

y 1 = rk4  y n f  y n x n  x n h  (5.9)

64 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Model Definition

Then estimate error:

y1 = y2 – y1 (5.10)

err =MAX  ABS  y 1 /yscale  / (5.11)

If the error is small (err <= 1.0) the function returns

y
y n + 1 = y 2 + -----1- (5.12)
15

which advances a solution y from xn to xn+1 = xn + h

or else the time step is reduced and the function tries again.

5.1.4 Update frequency

The time step for the update of the MIKE ECO Lab equations is the overall
time step (specified on the Time dialog) multiplied by the Update Frequency.

Selecting the time step of the MIKE ECO Lab model, and hereby the Update
Frequency, has to be based on considerations of the time scales of the pro-
cesses involved. Please notice that this selection can be rather decisive for
the precision of the numerical solution as well as for the CPU time of the sim-
ulation. A large Update Frequency will decrease the precision as well as the
CPU time. It is therefore advisable to perform a sensitivity analysis on the
Update Frequency before making the final selection.

The purpose of not updating the MIKE ECO Lab every time step is to reduce
simulation time.

Factors influencing the CPU time

The CPU time required by a hydrodynamic and MIKE ECO Lab simulation
depends on the size of your model, on the number of time steps in your simu-
lation, on the complexity of the used MIKE ECO Lab template, on the chosen
numerical solution method and on the general computational speed of your
computer.

If you wish to estimate how changes in your MIKE ECO Lab set-up changes
the CPU time required, the following guidelines can be used:

 the CPU time varies linearly with the number of water points (or compu-
tational points) in the model.
 the CPU time also varies linearly with the number of time steps.
 the CPU time increases approximately linearly with the number of advec-
tive state variables.

Powering Water Decisions 65

 MIKE ECO LAB / OIL SPILL MODULE

Be aware that the following also influence CPU time:

 the choice of MIKE ECO Lab integration method

 the number and complexity of the mathematical expressions in the MIKE
ECO Lab template
 the MIKE ECO Lab update frequency

5.1.5 Options

Compute in dry elements

If this option is enabled, ECO Lab is run on all elements, regardless of their
flood/dry status. For performance, ECO Lab calculations are usually just per-
formed on "real wet" elements, similar to AD transport calculations, i.e. dry
elements are skipped. Enabling the calculation on dry elements will allow you
to calculate fixed (not transported) ECO Lab items, i.e. fixed state variables.
The concentrations of any transported state variable/ AD component will be
set to zero in all elements considered dry and not be updated. Enabling the
dry element calculation will also cause the output of all fixed state variables in
dry elements. Please note that process/auxiliary variables are not written by
default in dry elements. If you need these, ensure that the output specification
uses the flood and dry type "whole area".

Optimise unused equations

Enabling this option may speed up the calculation time of ECO Lab. Before
the simulation start, the ECO Lab core will replace any calculation expression
of unused variables with a constant value, i.e. it will not waist time in calculat-
ing unnecessary expressions. "Unused" in this context means that the
expression is not direct or indirect relevant as a model result output and it has
no side effects(1). Please note that the runtime becomes dependent on the
output specifications.

Disable Process
This is an option considered relevant only during model development/setup. It
allows to fully skip all ECO Lab calculations, i.e. just the necessary AD trans-
port and utility calculations are performed. If enabled, the model results corre-
spond to a setup with the corresponding number of AD components.

5.2 Classes
The Classes dialog shows a summary of the classes defined in the selected
model definition. This dialog is only visible if the chosen template contains a
valid particle class description.

1 Side effects are indirect effects on the calculations. For example, the computation of random numbers has some
side effects. Skipping/removing one random number calculation would change the sequence of numbers created
by the pseudo-random generator and potentially cause a different simulation outcome. Similar, skipping a call to
an external plugin could result in unintended behavior. Therefore, expressions with these and a number of other
functions are not replaced/ optimised.

66 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

State Variables

The name and description of each class variable are given. From the list view
you can go to the dialog for specification by clicking on the “Go to..” button.

5.2.1 General description

Each particle class may be defined by a number of attributes like state varia-
bles, constants and restricted area search functions. State variables of a
class represent internal states and they are the variables the user wants usu-
ally to predict. Constants usually represent rates or thresholds whereas
restricted area search functions are used by a particle to interact/sense the
environment.

5.2.2 Class specification

This single class overview page is just a static information page listing the
class name and description and the number of pre-defined state variables,
constants, arithmetical and restricted area search functions of this specific
class.

State variables
The state variables dialog shows a summary of the state variables defined.
The state variables are static.

Constants
The class constants dialog lists all constants defined for this class with their
name, description and unit together with a user-defined numeric value. The
number and type of constants cannot be modified. If no class constants are
defined the dialog is not shown.

Restricted area search functions

The class restricted area search functions page lists all restricted area search
functions defined for this class with their number and description. The num-
ber and type of functions cannot be modified. If no spatial constants are
defined the dialog is not shown.

5.3 State Variables

The state variables dialog shows a summary of the state variables defined in
the MIKE ECO Lab model. The Symbol, Description, Unit and Transport type
("NONE" refers to a fixed State Variable and "AD" refers to a moveable state
variable, which is transported by Advection-dispersion) of each State Variable
are given. If no state variables are defined the dialog is not shown.

Powering Water Decisions 67

 MIKE ECO LAB / OIL SPILL MODULE

5.3.1 General description

State variables are usually the most important variables in a process oriented
MIKE ECO Lab template. State variables represent those variables that
describe the state of the ecosystem and that the user wants to predict the
state of. So they are also the main result after running a MIKE ECO Lab set-
up. The state variable will change according to how the processes that the
model developer describes affects it. So the critical task for the model devel-
oper is to describe the processes in a way, so that the state of the state varia-
ble changes correctly, also when forcings (external conditions such as e.g.
wind, temperature) change.

5.3.2 Transport

In MIKE ECO Lab the state variables that are present in the water column
has the option of using the advection-dispersion module for calculating the
transport based on hydrodynamics. If this option is not specified in the MIKE
ECO Lab template the state variable will have a fixed location and the calcu-
lation of its state will only be based on the processes defined in MIKE ECO
Lab template and not based on advection-dispersion processes.

State variables that are present in the sediment can not be transported based
on the advection-dispersion module.

It is also possible to have state variables in the water column that have a fixed
nature. An example of that could be relevant in models involving aquaculture,
for instance cages of scallops in the water column.

5.3.3 Units

In MIKE ECO Lab it is possible to use a limited number predefined unit types
that are supporting the EUM unit conversion system.

For state variables there are 3 unit types to choose from in the MIKE ECO
Lab template editor.

1. Volume Concentration [mg/l]. In the engine dialogs of MIKE ECO Lab

this unit can be converted to other units for mass and volume chosen by
the user.
2. Area Concentration [g/m2]. In the engine dialogs of MIKE ECO Lab this
unit can be converted to other units for mass and area chosen by the
user.
3. Undefined, here some user specified text can be written that will be writ-
ten in e.g. output, but no conversion will take place by the EUM unit con-
version system.

68 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Solution Technique

5.4 Solution Technique

The simulation time and accuracy can be controlled by specifying the order of
the numerical schemes which are used in the numerical calculations. Both
the scheme for time integration and for space discretization can be specified.
You can select either a lower order scheme (first order) or a higher order
scheme. The lower order scheme is faster, but less accurate. For more
details on the numerical solution techniques, see the scientific documenta-
tion. If no state variables are defined this dialog is not shown.

The time integration of the transport (advection-dispersion) equations is per-

formed using an explicit scheme. Due to the stability restriction using an
explicit scheme the time step interval must be selected so that the CFL num-
ber is less than 1. A variable time step interval is used in the calculation and it
is determined so that the CFL number is less than a critical CFL number in all
computational nodes. To control the time step it is also possible for the user to
specify a minimum time step and a maximum time step. The time step inter-
val for the transport equations is synchronized to match the overall time step
specified on the Time dialog.

The minimum and maximum time step interval and the critical CFL number is
specified in the Solution Technique dialog in the HYDRODYNAMIC MOD-
ULE.

5.4.1 Remarks and hints

If the important processes are dominated by convection (flow), then higher

order space discretization should be chosen. If they are dominated by diffu-
sion, the lower order space discretization can be sufficiently accurate. In gen-
eral, the time integration method and space discretization method should be
chosen alike.

Choosing the higher order scheme for time integration will increase the com-
puting time by a factor of 2 compared to the lower order scheme. Choosing
the higher order scheme for space discretization will increase the computing
time by a factor of 1½ to 2. Choosing both as higher order will increase the
computing time by a factor of 3-4. However, the higher order scheme will in
general produce results that are significantly more accurate than the lower
order scheme.

The default value for the critical CFL number is 1, which should secure stabil-
ity. However, the calculation of the CFL number is only an estimate. Hence,
stability problems can occur using this value. In these cases you can reduce
the critical CFL number. It must be in the range from 0 to 1. Alternatively, you
can reduce the maximum time step interval. Note, that setting the minimum
and maximum time step interval equal to the overall time step interval speci-
fied on the Time dialog, the time integration will be performed with constant
time step. In this case the time step interval should be selected so the the
CFL number is smaller than 1.

Powering Water Decisions 69

 MIKE ECO LAB / OIL SPILL MODULE

The total number of time steps in the calculation and the maximum and mini-
mum time interval during the calculation are printed in the log-file for the sim-
ulation. The CFL number can be saved in an output file.

The higher order scheme can exhibit under and over shoots in regions with
steep gradients. Hence, when the higher order scheme is used in combina-
tion with a limitation on the minimum and maximum value of the concentra-
tion, then mass conservation cannot be guarenteed.

5.5 Constants
In the MIKE ECO Lab Constants dialog the MIKE ECO Lab constants
defined in the MIKE ECO Lab template can be modified. If no constants are
defined the dialog is not shown.

The constants are defined as any input parameter (physical constant, coeffi-
cient, rate, etc.) in the MIKE ECO Lab model, which is constant in time. The
constants are essentially divided into two groups:

 Built-in constants
 User-specified Constants.

The built-in constants are automatically provided by the model system during
execution, whereas the user-specified constants have to be specified in the
present dialog. Depending on the spatial variation of the constant, as defined
in the MIKE ECO Lab template, it can be specified as a "Constant value", a
"Type 2 data file" or a dfsu file with one time step.

NOTE: A constant, which is defined as a built-in constant in the MIKE ECO

Lab template, will appear as a user-defined constant in case it is not sup-
ported by MIKE 21 Flow Model FM.

5.5.1 General description

Constants are used as arguments in the mathematical expressions of pro-

cesses in a MIKE ECO Lab template. They are constant in time, but can vary
in space. Typical examples of constants are different specific rate coefficients,
exponents, halfsaturation concentrations and also universal constants such
as gas constant and atomic weights.

5.5.2 Built-in constants

It is possible in MIKE ECO Lab templates to refer to constants that are speci-
fied or calculated in the hydrodynamic engines of MIKE. The available built-in
constants are not the same in all the hydrodynamic engines. If it is specified
that it should be a built-in constant and it is not available in the engine, the
engine dialog will prompt as if the constant was user specified.

70 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Forcings

It is possible to choose from the following list of built-in constants in MIKE 21

Flow Model FM:

1. Latitude. If available, this built-in constant returns the latitude of the

model area.
2. Longitude. If available, this built-in constant returns the longitude of the
model area.
3. Bed level. If available, this built-in constant returns the altitude of the
model area.
4. AD time step (AD_DT). This built-in constant returns the duration of time
between every advection-dispersion calculation.
5. MIKE ECO Lab time step (WQ_DT). This built-in constant returns the
time step in seconds between every update by the MIKE ECO Lab Mod-
ule.

5.5.3 Remarks and hints

Some coefficients and transformation rates can be found in the litterature, but
many have not been documented in litterature. So this possible lack of infor-
mation is wise to investigate, before defining a complex MIKE ECO Lab
model.

5.6 Forcings
In the MIKE ECO Lab Forcings dialog the MIKE ECO Lab forcings defined in
the MIKE ECO Lab template can be modified. The Forcings are defined as
any input parameter (physical property, rate, etc.) in the MIKE ECO Lab
model, which is varying in time. Examples of a forcing are: Temperature,
salinity, solar radiation, and water depth. The forcings are essentially divided
into two groups:

 Built-in forcings
 User-specified forcings.

The built-in forcings are automatically provided by the model system during
execution, whereas the user-specified forcings have to be specified in the
present dialog. Depending on the spatial variation of the forcing, as defined in
the MIKE ECO Lab model, it can be specified as a "Constant value", a "Type
0 data file", a "Type 2 data file" or a dfsu file. The data files have to cover the
entire simulation period. In case a type 2 file is allowed for, it must cover the
spatial extent and coordinates of the mesh file.

NOTE: A forcing, which is defined as a built-in forcing in the MIKE ECO Lab
model, will appear as a user-defined forcing here in case it is not supported
by MIKE 21 Flow Model FM.

Powering Water Decisions 71

 MIKE ECO LAB / OIL SPILL MODULE

5.6.1 General description

Forcings are used as arguments in the mathematical expressions of pro-

cesses in a MIKE ECO Lab template. They can vary in time and space. They
represent variables of an external nature that affect the ecosystem. Typical
examples of forcings are temperature, solar radiation and wind.

5.6.2 Built-in forcings

It is possible in MIKE ECO Lab templates to refer to forcings that are speci-
fied or calculated in the hydrodynamic engines of MIKE. The available built-in
forcings are not the same in all the DHI hydrodynamical engines. If it is spec-
ified that it should be a built-in forcing and it is not available in the engine, the
engine dialog will prompt as if the forcing was user specified.

It is possible to choose from the following list of built-in forcings:

1. Water layer height. This forcing returns the spatial distance between lay-
ers. In one layered grid systems the water depth will be returned
2. Salinity. If available, this forcing returns the salinity calculated in the
hydrodynamic set-up
3. Temperature. If available, this forcing returns the temperature calculated
in the hydrodynamic set-up
4. Wind velocity. If available, this forcing returns the wind velocity specified
in the hydrodynamic set-up
5. Wind direction. If available, this forcing returns the wind direction speci-
fied in the hydrodynamic set-up
6. Water surface level. This forcing returns the water level calculated in the
hydrodynamic set-up
7. Water depth. This forcing returns the water depth calculated in the hydro-
dynamic set-up
8. Horizontal current speed. This forcing returns the horizontal current
speed calculated in the hydrodynamic set-up
9. Horizontal current direction. This forcing returns the horizontal current
direction calculated in the hydrodynamic set-up
10. Vertical current speed. This forcing returns the vertical current speed cal-
culated in the hydrodynamic set-up
11. Water surface slope in flow direction. This forcing returns the slope of the
water surface calculated in the hydrodynamic set-up. Under normal con-
ditions positive values.
12. Water density. If available in hydrodynamic simulation this forcing returns
a value for water density

72 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Dispersion

5.6.3 Remarks and hints

If many measurements are available of a state variable, it is possible to tem-

porarily replace it with a forcing. This way a specific process can be isolated
and calibrated with less uncertainty. After calibrating the process, the forcing
can be transformed back into a state variable again.

5.7 Dispersion
In numerical models the dispersion usually describes transport due to non-
resolved processes. In coastal areas it can be transport due to non-resolved
turbulence or eddies. Especially in the horizontal directions the effects of non-
resolved processes can be significant, in which case the dispersion coeffi-
cient formally should depend on the resolution.

The dispersion is specified individually for each AD state variable and for
each particle class.

5.7.1 Dispersion specification

The dispersion can be formulated in one of three ways

 No dispersion
 Dispersion coefficient formulation
 Scaled eddy viscosity formulation

Selecting the dispersion coefficient formulation you must specify the disper-
sion coefficient.

Using the scaled eddy viscosity formulation the dispersion coefficient is cal-
culated as the eddy viscosity used in solution of the flow equations multiplied
by at scaling factor. For specification of the eddy viscosity see section 6.6,
Eddy Viscosity.

Data
Selecting dispersion coefficient option the format of the dispersion coefficient
can be specified as

 Constant (in both time and domain)

 Varying in domain

For the case with dispersion coefficient varying in domain you have to pre-
pare a data file containing the dispersion coefficient before you set up the
hydrodynamic simulation. The file must be a 2D unstructured data file (dfsu)
or a 2D grid data file (dfs2). The area in the data file must cover the model

Powering Water Decisions 73

 MIKE ECO LAB / OIL SPILL MODULE

area. If a dfsu-file is used, a piecewise constant interpolation is used to map

the data. If a dfs2-file is used, a bilinear interpolation is used to map the data.

Selecting Scaled eddy viscosity option the format of the scaling factor can be
specified as

 Constant
 Varying in domain

For the case with values varying in domain you have to prepare a data file
containing the scaling factor before you set up the hydrodynamic simulation.
The file must be a 2D unstructured data file (dfsu) or a 2D data grid file (dfs2).
The area in the data file must cover the model area. If a dfsu-file is used, a
piecewise constant interpolation is used to map the data. If a dfs2-file is used,
a bilinear interpolation is used to map the data.

5.7.2 Recommended values

When more sophisticated eddy viscosity models are used, as the Smagorin-
sky or k- models, the scaled eddy formulation should be used.

The scaling factor can be estimated by 1/T, where T is the Prandtl number.
The default value here for the Prandtl number is 0.9 corresponding to a scal-
ing factor of 1.1.

The dispersion coefficient is usually one of the key calibration parameters for
the Transport Module. It is therefore difficult to device generally applicable
values for the dispersion coefficient. However, using Reynolds analogy, the
dispersion coefficient can be written as the product of a length scale and a
velocity scale. In shallow waters the length scale can often be taken as the
water depth, while the velocity scale can be given as a typical current speed.

Values in the order of 1 are usually recommended for the scaling factor. For
more information see e.g. (Rodi, 1980 /2/).

5.8 Drift profile

This section is only relevant when including particle classes.

5.8.1 General description

The drift profile is a description of the drift regime that the particles are influ-
enced by. It will normally be the currents and the wind that governs the drift
regime. The currents including wind are already calculated in the hydrody-
namic setup, but the hydrodynamic output does normally not have a fine dis-
cretization of the current profile near the bed. For MIKE 21 FM for instance it
is depth average values that are the output of the hydrodynamic setup, and

74 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Drift profile

even for MIKE 3 FM the drift conditions near the bed are normally not
resolved adequately for describing sediment particles. Therefore it can be a
good idea to assume some shapes of the vertical drift profile.

It is possible to include the bed friction drag on the current profile by assum-
ing a bed shear profile (or logarithmic profile). For MIKE 21 FM the bed shear
profile will be applied in the whole water column.

The wind drag can also cause increased flow velocities in the upper part of
the water column, and corresponding velocities in the opposite direction in
the lower part. In MIKE 3 FM this effect can be included in the hydrodynamic
output, but the depth averaged MIKE 21 FM is not able to do that. So if this
flow regime should be described in MIKE 21 FM PT, the wind induced profile
must be applied, that will distribute the depth averaged flow in the water col-
umn.

If particles are in the water surface, they can be influenced directly by the
wind additionally to the influence from the flow. This will result in an additional
acceleration of the particle in a direction relative to the wind turned with a
wind drift angle caused by coriolis forces.

5.8.2 Class drift profile

The different drift profile types can be combined in the following ways

 Use raw data from hydrodynamics

 Use bed shear profile
 Use surface wind acceleration
 Use bed shear profile and wind induced profile
 Use bed shear profile, wind induced profile and surface wind accelera-
tion
 Use bed shear profile and surface wind acceleration

Depending on the choice of drift profile you specify some additional parame-
ters.

Bed shear profile

The shape of the velocity profile within a turbulent boundary layer is well
established by both theory and experience. The profile has specific charac-
teristics very close to the bed where viscosity controls the vertical transport of
momentum, and different characteristics farther from the bed where turbu-
lence controls the vertical transport of momentum. The region closest to the
bed boundary is called the laminar sub-layer or viscous sub-layer, because
within the region turbulence is suppressed by viscosity. The laminar sublayer
only plays a significant role for smooth flows, whereas for rough flows the
flow is zero for z smaller than z0.

Powering Water Decisions 75

 MIKE ECO LAB / OIL SPILL MODULE

Logarithmic layer (z >= s smooth flow) (z >= z0 rough flow)

2,3 x
u = -------  u *  log 10 ----- (5.13)
k zo

where

u*: Friction velocity

k: Karman constant = 0.4
z0 : characteristic roughness

Surface wind acceleration

The user must specify wind weight, wind drift angle and the kinematic viscos-
ity in order to calculate the surface wind acceleration.

Particles that are exposed to wind in the water surface are affected according
to the wind regime in 2 ways: Indirectly via the currents that include the wind,
but also directly as an extra force on the particle. How much of the wind
speed that are transferred to the particle speed depends on the nature of the
particle, how much is the particle exposed etc, so therefore it is a calibration
factor that expresses how much of the windspeed that is added to the particle
speed.

U particle = U current + C w  W  sin   dir –  +  w 

(5.14)
V particle = V current + C w  W  cos   dir –  +  w 

where

Cw : Calibration factor for wind drag on particle

dir: Wind direction
w: Wind drift angle

Wind drift angle

The coriolis force is normally included in the hydrodynamic currents, but for
the wind acceleration of surface particles the coriolis force must be consid-
ered as well.

Due to the influence from the Coriolis force, the direction of the wind drift vec-
tor is turned relatively to the wind direction. The angle qw of deviation is
termed with the wind drift angle. It turns to the right on the Northern Hemi-
sphere and to the left on the Southern Hemisphere. Thus wind drift angles for
the Southern Hemisphere should be specified by negative values. From Al-
Rabeh (1994), it is assumed that

 U w 3
 w = exp  ----------------
- (5.15)
 g w 

76 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Drift profile

where

: -0.3.10-8
: 2838’
w: Kinematic viscosity [m2/s]
g: Gravity [m/s2]

The magnitude of the wind drift angle varies with the geographical location
and wind speed and it is often estimated at 12-15 degrees in the North Sea.

Wind induced profile

The user must specify a wind drift factor, a depth of influence and an offshore
limit in order to calculate a wind induced drift profile.

The direction and the magnitude of currents are varying with the depth. The
wind may induce strong currents in the surface layer and has a significant
effect on the surface drifting.

The depth-averaged currents must therefore be modified with respect to the

influence from the wind. This has been done by calculating the tracking as a
vectorial sum of the wind driven currents (the wind drift vector) and the cur-
rent velocity vector.

The wind velocity component is added to the current velocity component vec-
torially to determine the total surface drift velocity. Once the drift velocity is
known the movement of the parcels are computed by numerical integration.

The magnitude of the wind drift vector is commonly assumed to be propor-

tional to the magnitude of the wind speed 10 m above the sea surface.

The vertical distribution of the wind drift vector consists of an offshore part
and an onshore part. The onshore distribution is based on a parabolic vertical
profile and is able to produce back flow at depth, where the offshore logarith-
mic profile does not.

The parabolic profile acts in shallow waters with a water depth less than a
specified water depth, hsep, which is a positive value and measured from the
free water surface.

The vertical distribution of the parabolic onshore profile is given by

z
c w  z  = c w*  1 – 3 ---  1 – --z- (5.16)
 h  h

where

h: local water depth

z: vertical particle co-ordinate, measured from sea surface
cw*: wind drift factor (input)

Powering Water Decisions 77

 MIKE ECO LAB / OIL SPILL MODULE

The parabolic profile causes the wind-generated flow in the upper third of the
water column to be in the same direction as the current and the flow in the
lower part to be in the opposite direction of the wind. There is no net depth
averaged mass transport due to the wind.

The vertical distribution of the offshore wind drift vector is given by

c w  z  = c w* exp  – k 0 z  (5.17)

where

k0 : 3/hw
hw: depth of wind influence (input)
z: vertical co-ordinate measured from sea surface
cw*: wind drift factor [-] (input)

5.9 Bed Roughness

This section is only relevant when including particle classes.

The bed roughness is used in the particle transport sub-model if the class drift
profile is defined as a bed shear profile (logarithmic profile). It is an important
parameter for describing the near bed flow conditions.

The bed roughness can be specified as

 Bed roughness from hydrodynamic model

 User specified bed roughness

5.9.1 General description

The bed roughness in the particle transport sub-model is defined by the

Nikuradse roughness ks. This roughness parameter may be converted from
other roughness definitions as described below.

Chezy number conversion to Nikuradse roughness ks:

12h -
k s = ------------------ (5.18)
 C  18 
10

where

h: Water depth
C: Chezy number

78 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Wind Forcing

Manning number conversion to Nikuradse roughness ks:

6
25,4
k s =  ---------- (5.19)
M

where M is the Manning number.

5.9.2 Bed roughness from hydrodynamic model

The bed resistance is often already specified in the hydrodynamic setup.

In MIKE 21 FM bed resistance can be specified as Manning number, or

Chezy number. If bed roughness from hydrodynamic model is selected, the
Manning number, or Chezy number will be automatically converted to
Nikuradse roughness ks (see 5.9.1 General description).

5.9.3 User specified bed roughness

The bed roughness must be specified as Nikuradse roughness and it can be

specified as

 Constant
 Varying in domain

For the case with values varying in domain you have to prepare a data file
containing the bed roughness before you set up the hydrodynamic simula-
tion. The file must be a 2D unstructured data file (dfsu) or a 2D grid data file
(dfs2). The area in the data file must cover the model area. If a dfsu-file is
used, a piecewice constant interpolation is used to map the data. If a dfs2-file
is used, a bilinear interpolation is used to map the data.

5.10 Wind Forcing

This section is only relevant when including particle classes.

The wind forcing is used in the particle transport sub-model if the Class drift
profile is defined using the influence from wind.

The wind forcing can be specified as

 User specified wind, speed and direction

 User specified wind and velocity components
 Wind from hydrodynamic model

Powering Water Decisions 79

 MIKE ECO LAB / OIL SPILL MODULE

5.10.1 User specified wind

The wind forcing can be specified as

 Constant
 Varying in time, constant in domain
 Varying in time and domain

For the case with values varying in domain you have to prepare a data file
containing the wind forcing (speed and direction (in degrees from true North)
or velocity components) before you set up the hydrodynamic simulation. The
file must be a 2D unstructured data file (dfsu) or a 2D grid data file (dfs2). The
area in the data file must cover the model area. If a dfsu-file is used, a piece-
wice constant interpolation is used to map the data. If a dfs2-file is used, a
bilinear interpolation is used to map the data.

Furthermore a soft start interval must be specified.

5.11 Precipitation-Evaporation
If your simulation includes precipitation and/or evaporation, you need to spec-
ify the concentration of each AD state variable in the precipitated and evapo-
rated water mass. The precipitation and evaporation can be included in two
ways

 Ambient concentration.
The concentration of the precipitated/evaporated water mass is set equal
to the concentration of the ambient sea water.
 Specified concentration.
The concentration of the precipitated/evaporated water mass is specified
explictly.

If you have chosen the net precipitation option in the Precipitation-Evapora-

tion dialog in the HYDRODYNAMIC MODULE, the precipitation concentration
will be used when the specified net-precipation is positive. When the net-pre-
cipitation is negative, the evaporation concentration is used.

The precipitation-evaporation is specified individually for each AD state varia-

ble.

5.11.1 Specification

Data
Selecting the specified concentration option the format of the concentration
(in component unit) can be specified as

80 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Infiltration

 Constant (in both time and domain)

 Varrying in time and constant in domain
 Varying in both time and domain

For the case with concentration varying in time but constant in domain you
have to prepare a data file containing the concentration (in component unit)
before you set up the hydrodynamic simulation. The data file must be a time
series file (dfs0). The data must cover the complete simulation period. The
time step of the input data file does not, however, have to be the same as the
time step of the hydrodynamic simulation. A linear interpolation will be
applied if the time steps differ.

For the case with concentration varying both in time and domain you have to
prepare a data file containing the concentration (in component units) before
you set up the hydrodynamic simulation. The file must be a 2D unstructured
data file (dfsu) or a 2D grid data file (dfs2). The area in the data file must
cover the model area. If a dfsu-file is used, a piecewice constant interpolation
is used to map the data. If a dfs2-file is used, a bilinear interpolation is used to
map the data. The data must cover the complete simulation period. The time
step of the input data file does not, however, have to be the same as the time
step of the hydrodynamic simulation. A linear interpolation will be applied if
the time steps differ.

Soft start interval

You can specify a soft start interval during which the precipitation/evaporation
concentration is increased linearly from 0 to the specified values of the pre-
cipitation/evaporation concentration. By default the soft start interval is zero
(no soft start).

5.11.2 Recommended values

Usually it is most sensible to set the concentration of the evaporated water

mass to zero, in which case the component is conserved in the water.

5.12 Infiltration
If your simulation includes infiltration you need to specify the concentration of
each AD state variable in the infiltrated water mass. This can be done two
ways

 Ambient concentration.
The concentration in the infiltrated water mass is set equal to the con-
centration in the ambient sea water.
 Specified concentration.
The concentration in the infiltrated water mass is specified explictly.

Powering Water Decisions 81

 MIKE ECO LAB / OIL SPILL MODULE

Data
Selecting the specified concentration option the format of the concentration
can be specified as:

 Constant (in both time and domain)

 Varying in time and constant in domain
 Varying in time and domain

For the case with concentration varying in time but constant in domain you
have to prepare a data file containing the concentration (in component unit)
before you set up the hydrodynamic simulation. The data file must be a time
series file (dfs0). The data must cover the complete simulation period. The
time step of the input data file does not, however, have to be the same as the
time step of the hydrodynamic simulation. A linear interpolation will be
applied if the time steps differ.

For the case with concentration varying both in time and domain you have to
prepare a data file containing the concentration (in component units) before
you set up the hydrodynamic simulation. The file must be a 2D unstructured
data file (dfsu) or a 2D grid data file (dfs2). The area in the data file must
cover the model area. If a dfsu-file is used piecewice constant interpolation is
used to map the data. If a dfs2-file is used bilinear interpolation is used to
map the data. The data must cover the complete simulation period. The time
step of the input data file does not, however, have to be the same as the time
step of the hydrodynamic simulation. A linear interpolation will be applied if
the time steps differ.

Soft start interval

You can specify a soft start interval during which the infiltrated concentration
is increased linearly from 0 to the specified values of the infiltrated concentra-
tion. By default the soft start interval is zero (no soft start).

5.13 Sources
Point sources of dissolved components are important in many applications as
e.g. release of nutrients from rivers, intakes and outlets from cooling water or
desalination plants.

In the MIKE ECO Lab Module, the source concentrations of each component
in every sources point can be specified. The number of sources, their generic
names, location and discharges magnitude are specified in the Sources dia-
log in the HYDRODYNAMIC MODULE.

Depending on the choice of property page you can see a geographic view or
a list view of the sources.

82 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Sources

The source concentrations are specified individually for each source and
each component. From the list view you can go to the dialog for specification
by clicking on the “Go to..” button.

5.13.1 Source specification

The type of sources can be specified in two ways

 Specified concentration
 Excess concentration

The source flux is calculated as Qsource.Csource where Qsource is the magnitude

of the source and Csource is the concentration of the source. The magnitude of
the source is specified in the Sources dialog in the HYDRODYNAMIC MOD-
ULE.

When selecting the specified concentration option the source concentration is

the specified concentration if the magnitude of the source is positive (water is
discharged into the ambient water). The source concentration is the concen-
tration at the source point if the magnitude of the source is negative (water is
discharge out the ambient water). This option is pertinent to e.g. river outlets
or other sources where the concentration is independent of the surrounding
water.

When selecting the excess concentration option the source concentration is

the sum of the specified excess concentration and concentration at a point in
the model if the magnitude of the source is positive (water is discharge into
the ambient water). If it is an isolated source the point is the location of the
source. If it is a connected source, the point is the location where water is dis-
charged out of the water. The source concentration is the component concen-
tration at the source point if the magnitude of the source is negative (water is
discharged out of the ambient water).

Data
The format of the source information can be specified as

 Constant (in time)

 Varying in time

For the case with source concentration varying in time you have to prepare a
data file containing the concentration (in concentration units) of the source
before you set up the hydrodynamic simulation. The data file must be a time
series file (dfs0). The data must cover the complete simulation period. The
time step of the input data file does not, however, have to be the same as the
time step of the hydrodynamic simulation. A linear interpolation will be
applied if the time steps differ.

Powering Water Decisions 83

 MIKE ECO LAB / OIL SPILL MODULE

5.13.2 Remarks and hints

Point sources are entered into elements, such that the inflowing mass of the
component initially is distributed over the element where the source resides.
Therefore the concentration seen in the results from the simulation usually is
lower than the source concentration.

5.14 Particle Sources

Particle sources are only available when the chosen template defines at least
one valid particle class. A particle source is used to introduce members or
particles of a specific class into the model. Particle sources do not depend on
water inflow and thus must be specified independently of hydraulic sources.

By switching property page you can see a Geographic View or a List View of
the sources. New sources can be added to both these views.

In the List View you can create a new source by clicking on the "New source"
button. By selecting a source in the Source list and clicking on the "Delete
source" you can remove this source. For each source you can specify the
name of the source and whether the source should be active or not. The
specification of detailed information for each source is made subsequently.
From the List View page you can go to the dialog for specification by clicking
on the "Go to .." button.

In the Geographic View it is also possible to create a new source by selecting

“Add new...” from the context menu. The geographical position of the source
is set to the mouse pointer coordinates and depending on the source type,
the source location can then be edited by dragging the source symbols on the
map. The additional information for the source is made subsequently.

5.14.1 Particle source specification

Three general types of source layouts can be specified:

 Point source
 Area source
 File Source

Each of the source layouts has several sub-types. In general, point sources
specify a single, defined point in space whereas area sources describe a spa-
tial domain where particles are released. In case of a file source the release
locations and particle states are read from a file. A normal source is active
throughout the complete simulation time. An initial source is used to specify
an initial particle distribution and does not release any particles during the
rest of simulation.

84 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Particle Sources

A source may include one or many different types of particle classes. The
additional information for each particle class is made subsequently. You may
choose to include or exclude the defined classes in the simulation.

Coordinates
You have to specify the map projection (Longitude/Latitude, UTM, etc.) in
which you want to specify the location coordinates of the particle source. In
case of a file source the release coordinates are read from a data file.

Vertical source specification

Apart from the horizontal specification, a source location must be specified
vertically in the water column. Three definitions are available for the vertical
positioning of a source. The vertical reference can be defined as :

 Depth (positive downwards from the moving water surface)

 Above bed (positive upwards relative to the sea bed)
 Datum (absolute level)

An area source may further spawn over a range in the water column. In this
case a vertical layer thickness has to be given. The vertical position of a
released particle is then randomly selected within the defined layer, i.e.
extending from the reference point plus the layer thickness. All data on verti-
cal position refer to the length unit. If the source definition causes a particle to
be released above surface/below bed level, the position is set to the sur-
face/bed respectively. In case of a file source the vertical position is read from
the data file.

Point source
The location of a point source must be defined using one of two sub-types

 Fixed location
 Moving location

For the fixed location source option , i.e. the source is stationary during simu-
lation time, you must specify the horizontal coordinates of the source point
and its vertical position.

For the moving location source option, i.e. the source is moving along a
defined path, you must specify a time series file (dfs0) that contains the hori-
zontal coordinates and the vertical position of the source as a function of
time. The unit of the vertical position is meters and the value is to be given
according to the defined vertical type.

Area source
The location of an area source must be defined using one of two sub-types

 Point collection

Powering Water Decisions 85

 MIKE ECO LAB / OIL SPILL MODULE

 Contour/Shape

An area source describes either a collection of individual (fixed) points or a

geometric shape. A point collection can be an irregular placed group of indi-
vidual points or a regular grid within a defined rectangle. When declaring a
contour/shape source particles are released in the space covered by a
geometric figure. In all cases the specifications of the released particles is
valid for any location within the source.

Particle casting
Particles are placed within the area covered by the area source definition. It
may be that the described area includes land elements (e.g. a small island,
land borders etc.). In this case a particle potentially can be placed outside the
domain. If this happens, the particle may be re-casted to find a valid location
in the domain a given number of times. If this finally does not succeed, the
particle may be ignored (skip to the next particle) or the simulation will be
stopped with an error. Note that the amount of released substances will prob-
ably not match your expectations if particles are ignored as the particle asso-
ciated items will not enter the simulation!

Point collection
Point collections describe a number of individual, fixed source locations in the
horizontal plane belonging to the same source. Such a collection is defined
by one of two area definitions

 Regular grid
(points located at the cell centers in a given rectangular grid)
 Irregular point collection
(irregular placed, individual points)

The advantage over individual fixed point sources is that the item definitions
refer all release points.

Particle placing
Depending on the grid type the release locations will be used sequential (row
by column, column by row) or randomly permuted. Note that all positions
inside a point collection have to be used before a particle can be released at
the same position again, e.g. if there are 15 points in total and 15 particle per
time step are released, every location will receive one particle. Thus even if
random filling is selected all release locations will be used before a new per-
mutation is selected.

Regular grid
In case of a regular grid the anchor point describes the lower left corner of the
rectangle covering the source area. The source rectangle extends a given
width along the X-axis and a height along the Y-axis and may be rotated
around the anchor point. The area is divided into a regular grid of N x M col-

86 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Particle Sources

umns and rows. The actual particle release positions are located at the cross-
ing/centres of the columns and rows.

Irregular point collection

An irregular grid consists of a number of individual, irregular placed points in
the horizontal plane.

Contour/Shape
A contour/shape source is defined by a set of points in the horizontal plane
and a number of elements made of these points to describe the actual
release area. Every release element must be characterised by three points.
Depending on the selected points the release location will be a single location
(all points of the element are the same), a line (2 points are the same) or a tri-
angle (three different points). The final release location is a random point on a
line element or inside a triangle element. Both the points and elements have
to be created manually and the first point must be placed inside the domain
(water point). Subsequent points may be placed outside water elements.

Particle placing
Normally particle placing is based on the total area. In this way particles cover
the total source area (covered by all elements) uniformly. Please note that by
definition elements describing single locations (all points in the element are
the same) or elements describing a line (two points are the same) do not
have an area. If particles are placed based on the total area, such elements
will not receive any particles! In this case the particles should be placed
based on the sub-element count. By doing this, every sub element has the
same probability to receive a particle.

Released class items

The number of particles to be released into the flow per overall time step can
be a constant number (same number every time step) or the number of
released particles can be read from a time series data file (dfs0). Be aware
that interpolation between data points may lead to an unexpected number of
particles, thus it is best to provide a data series with the same time resolution
than the simulation.

The individual state variables/class items can be specified separately. In gen-

eral the release format for a state variable can be constant, time varying or
from a specified distribution. The source data will be checked against the
allowed minimum/maximum item value. If the source value is outside this
range, the value will be clipped to ensure that the item value will not violate
the minimum/ maximum.

Certain state variables, e.g. a length or a gender indicator, etc., have to be

declared on a per-particle base. However, mass items may be declared as
the overall emitted amount, i.e. the amount is split among all released parti-
cles. This is especially useful if you change the number of particles as the

Powering Water Decisions 87

 MIKE ECO LAB / OIL SPILL MODULE

released amount will be independent on the number of particles. On the other

hand, the released amount can be specified on a per-particle base.

Similarly the amount of mass items can be declared to be handled as flux or

instantaneous release.

NOTE: The interpretation of the values depends on the combined settings for
the mass components and number of release particles.

Individual state variables

Each state variable/ class item can be configured separately. The format can
be either

 Constant
 Varying in time
 Within range
 Normal distribution
 Discrete distribution

If the format is set to constant or time varying, every particle will receive the
specified values, in case of a time series read from a time series data file
(dfs0). If the format is one of the following settings, each individual particle will
receive an individual value.

Within range
Value uniformly distributed between given Min and Max values.

Normal distribution
Value from a normal distribution with given mean and standard deviation.

Discrete distribution
Value determined from the specified discrete distribution.
A discrete distribution is characterised by a number of bins. Each bin has a
certain width and an associated value. The “width” determines the “propor-
tion” or probability for this bin and that a particle will receive the associated
value. The width can be a direct scale (i.e. number of items in that bin) or a
probability. The value has to be given in the correct item unit.

File Source
Currently only initial sources are supported as file source. In this case the file
source specification acts as a hotstart file source, i.e. particle count and coor-
dinates/states are read from a file. This information is used to generate an ini-
tial particle distribution where all particle states from a previous run are
restored from a data file. Therefore, besides the data file specification, no
other user input is required. Valid data files must contain all internal particle
states and can be generated by saving particle data as hotstart track. If a hot-
start track contains just one single timestep the date setting of the file is irrel-

88 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Initial Conditions

evant. If it contains more than one time step the stored period must cover the
whole simulation period. The storage frequency should match the simulation
time step, as no data interpolation is performed. In case of a mismatch, the
last matching time step in data file is used what could let to unexpected
results.

5.15 Initial Conditions

The initial conditions are the spatial distribution of the component concentra-
tion throughout the computational domain at the beginning of the simulation.
Initial conditions must always be provided. The initial conditions can be the
result from a previous simulation in which case the initial conditions effec-
tively act as a hot start of the concentration field for each state variable.

It is currently not possible to specify initial conditions for particle classes. Ini-
tial conditions can only be specified for normal state variables.

The initial conditions are specified individually for each state variable.

Data
The format of the initial concentration (in component unit) for the state varia-
ble can be specified as

 Constant (in domain)

 Varying in domain

For the case with varying in domain you have to prepare a data file containing
the concentration (in component unit) before you set up the hydrodynamic
simulation.The file must be a 2D unstructured data file (dfsu) or a 2D grid
data file (dfs2). The area in the data file must cover the model area. If a dfsu-
file is used, a piecewice constant interpolation is used to map the data. If a
dfs2-file is used, a bilinear interpolation is used to map the data. In case the
input data file contains a single time step, this field is used. In case the file
contains several time steps, e.g. from the results of a previous simulation, the
actual starting time of the simulation is used to interpolate the field in time.
Therefore the starting time must be between the start and end time of the file.

5.16 Boundary Conditions

Boundary conditions are only valid for AD-state variables (concentrations). If
a particle leaves the domain it is lost and can not re-enter the simulation. Ini-
tially, the set-up editor scans the mesh file for boundary codes (sections), dis-
plays the recognized codes and suggest a default name for each.You can re-
name these names to more meaningful names in the Domain dialog.

Depending on the choice of property page you can get a geographic view or a
list view of the boundaries.

Powering Water Decisions 89

 MIKE ECO LAB / OIL SPILL MODULE

The specification of the individual boundary information for each code (sec-
tion) and each component is made subsequently. From the list view you can
go to the dialog for specification by clicking the “Go to..” button.

5.16.1 Boundary specification

You can choose between the following three boundary types

 Land
 Specified values (Dirichlet boundary condition)
 Zero gradient (Neumann boundary condition).

Data
If specified values (Dirichlet boundary condition) is selected, the format of the
concentration (in component unit) for the AD state variable at the boundary
can be specified as

 Constant (in time and along boundary)

 Varying in time and constant along boundary
 Varying in time and along boundary

For the case with boundary data varying in time but constant along the
boundary you have to prepare a data file containing the concentration (in
component unit) before you set up the hydrodynamic simulation. The data file
must be a time series file (dfs0). The data must cover the complete simulation
period. The time step of the input data file does not, however, have to be the
same as the time step of the hydrodynamic simulation. You can choose
between different types of interpolation (see Interpolation type).

For the case with boundary data varying both in time and along the boundary
you have to prepare a data file containing the concentration (in component
unit) before you set up the hydrodynamic simulation. The data file must be a
profile file (dfs1). The data must cover the complete simulation period. The
time step of the input data file does not, however, have to be the same as the
time step of the hydrodynamic simulation. You can choose between different
types of time interpolation.

Interpolation type
For the two cases with values varying in time two types of time interpolation
can be selected:

 Linear
 Piecewise cubic

90 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Hydrodynamic Conditions

In the case with values varying along the boundary two methods of mapping
from the input data file to the boundary section are available:

 Normal
 Reverse order

Using normal interpolation, the first and last point of the line are mapped to
the first and the last node along the boundary section and the intermediate
boundary values are found by linear interpolation. Using reverse order inter-
polation, the last and first point of the line are mapped to the first and the last
node along the boundary section and the intermediate boundary values is
found by linear interpolation.

Soft start interval

You can specify a soft start interval during which boundary values are
increased from a specified reference value to the specified boundary value in
order to avoid shock waves being generated in the model. The increase can
either be linear or follow a sinusoidal curve.

5.17 Hydrodynamic Conditions

If the Hydrodynamic Module is not included in the simulation you have to pro-
vide the flow information using these settings. The MIKE ECO Lab module
will then run in a so called stand-alone mode, using an external flow field. The
flow information can be read from both unstructured (dfsu) or gridded
(dfs2/dfs3) data files. The data files need to cover the model area and span
over the entire simulation period. The flow information will be interpolated into
the domain mesh, i.e. it can be based on a different (coarser) resolution than
the actual domain mesh. If a dfsu-file is used, a piecewise constant interpola-
tion is used to map the data. If a dfs2-file is used, a bilinear interpolation is
used to map the data

2D setup (Mike21)
For a 2D setup you have to provide a 2D/area data file with the information on
either the surface level or the total water depth and the U- and V-components
of the flow field. If the bathymetry is not constant during the simulation period
(i.e. morphological changes) the file also has to provide the still water depth.

5.18 Temperature
The temperature can be specified as

 Not included
The model assumes a default value for the whole model domain
 User specified temperature

Powering Water Decisions 91

 MIKE ECO LAB / OIL SPILL MODULE

 From hydrodynamic model

The data calculated in the hydrodynamic module is used directly
(This option is not available when running a stand-alone setup)

5.18.1 User specified temperature

With a user specified setting, the data can be:

 Constant
 Varying in time, constant in domain
 Varying in time and domain

For the case with values varying in domain you have to prepare a data file
containing the information before you set up the hydrodynamic simulation.
The file must be a 2D unstructured data file (dfsu) or a 2D grid data file (dfs2).
The area in the data file must cover the model area. If a dfsu-file is used, a
piecewise constant interpolation is used to map the data. If a dfs2-file is used,
a bilinear interpolation is used to map the data

5.19 Salinity
The salinity can be specified as

 Not included
The model assumes a default value for the whole model domain
 User specified salinity
 From hydrodynamic model
The data calculated in the hydrodynamic module is used directly
(This option is not available when running a stand-alone setup)

5.19.1 User specified salinity

With a user specified setting, the data can be:

 Constant
 Varying in time, constant in domain
 Varying in time and domain

For the case with values varying in domain you have to prepare a data file
containing the information before you set up the hydrodynamic simulation.
The file must be a 2D unstructured data file (dfsu) or a 2D grid data file (dfs2).
The area in the data file must cover the model area. If a dfsu-file is used, a
piecewise constant interpolation is used to map the data. If a dfs2-file is used,
a bilinear interpolation is used to map the data

92 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Density

5.20 Density
The density can be specified as

 Not included
The model assumes a default value for the whole model domain
 User specified density
 From hydrodynamic model
The data calculated in the hydrodynamic module is used directly
(This option is not available when running a stand-alone setup)

5.20.1 User specified density

With a user specified setting, the data can be:

 Constant
 Varying in time, constant in domain
 Varying in time and domain

For the case with values varying in domain you have to prepare a data file
containing the information before you set up the hydrodynamic simulation.
The file must be a 2D unstructured data file (dfsu) or a 2D grid data file (dfs2).
The area in the data file must cover the model area. If a dfsu-file is used, a
piecewise constant interpolation is used to map the data. If a dfs2-file is used,
a bilinear interpolation is used to map the data

5.21 Output
Standard data files with computed results from the simulation can be speci-
fied here. Because result files tend to become large, it is normally not possi-
ble to save the computed discrete data in the whole area and at all time
steps. In practice, sub areas and subsets must be selected.

In the main outputs dialog you can add a new output file by clicking on the
"New output" button. By selecting a file in the Output list and clicking on the
"Delete output" you can remove this file. For each output file you can specify
the name (title) of the file and whether the output file should be included or
not. The specification of the individual output files is made subsequently. You
can go to the dialog for specification by clicking on the "Go to .." button.
Finally, you can view the results using the relevant MIKE Zero viewing/editing
tool by clicking on the "View" button during and after the simulation.

5.21.1 Geographic view

This shows the geographical position of the output data.

Powering Water Decisions 93

 MIKE ECO LAB / OIL SPILL MODULE

5.21.2 Output Specification

For each selected output file the field type, the output format, the data type,
the treatment of flood and dry, the output file (name, location and file type)
and time step must be specified. Depending on the output format and data
type the geographical extend of the output data must also be specified.

Field type
For a 2D simulation 2D field parameters can be selected. If the model defini-
tion contains state variables, the mass budget for a domain and the discharge
through a cross section can also be selected. If the model definition includes
particle classes, particle track, particle hotstart data track and particle track
mass budget for can be selected as well.

Output format
The possible choice of output format depends on the specified field type.

For 2D field variables the following formats can be selected

 Point series. Selected field data in geographical defined points.

 Lines series. Selected field data along geographical defined lines.
 Area series. Selected field data in geographical defined areas.

If mass budget is selected for the field type, you have to specify the domain
for which the mass budget should be calculated. The file type will be a dfs0
file.

If discharge is selected for the field type, you have to specify the cross sec-
tion through which the discharge should be calculated. The file type will be a
dfs0 file.

If particle track is selected for the field type, you have to specify the domain
within which the particle tracks should be saved. The track can be written
using various track storage formats, see Particle file format page 96.

If particle track mass budget is selected for the field type, you have to specify
the domain for which the mass budget should be calculated. The file type will
be a dfs0 file.

If particle hotstart data track is selected for the field type, you have to specify
the domain within which the particle tracks should be saved. Hotstart data
tracks are intended to be used as data file for a file source (hotstart) (see File
Source, page 48) and contain additional internal data states compared to
standard particle tracks. The track can be written using various track storage
formats, see Particle file format page 96, with the constrains that the track for-
mat must be a readable format (currently .xml and .track).

94 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Output

Data type
You must specify the data type. You can select discrete values or interpolated
values. For point series and area series both options are available. For line
series only interpolated values is available. See also Table 5.1.

If "discrete values" is selected for the data type, the values written in the data
file are the cell-averaged values.

If "interpolated values" is selected for the data type, the values written in the
data file are determined by 2nd order interpolation. The element in which the
point is located is determined and the point value is obtained by linear inter-
polation using the vertex (node) values for the actual element. The vertex val-
ues are calculated from on the cell-averaged values using the pseudo-
Laplacian procedure proposed by Holmes and Connell (1989).

Note, that all adjacent elements, including dry elements, are considered in
the interpolation calculation.

Output file
A name and location of the output file must be specified along with the file
type. The file type depends on the Output format and the Data type as shown
in Table 5.1.

Vectors components and tensors components in a 2D dfsu file are given in

the projection coordinate system. In a dfs2 file the vector and tensor compo-
nents are given in the local coordinate system.

Table 5.1 List of file types for the output files

Output format Data type File type

Point series Discrete values dfs0

Interpolated values dfs0

Line series Interpolated values dfs1

Area series Discrete values dfsu, dfs2*

Interpolated values dfs2

Cross-section Not relevant dfs0

Domain series Not relevant dfs0

Powering Water Decisions 95

 MIKE ECO LAB / OIL SPILL MODULE

* For area series with discrete values selected for the data type the file type
dfs2 can only be specified when the Mesh and bathymetry is specified using
a bathymetry data file.

Table 5.2 List of tools for viewing, editing and plotting results

File type Viewing/editing tools Plotting tools

dfs0 Time Series Editor Plot Composer

dfs1 Profile Series Editor Plot Composer

Result Viewer

dfs2 Grid Series Editor Plot Composer

Data Viewer Result Viewer
Data Viewer

dfsu Data Viewer Data Viewer

Data Manager Result Viewer
Plot Composer (2D)

Particle file format

Particle tracks may be stored in various storage formats. Figure 5.3 lists an
overview of the available format options. Most formats store the complete
particle track data, i.e. coordinates and variable values for a time step. Some
formats store only selected information like particle counts or statistics. Track
formats can in general be used by the MIKE software to visualise and post-
process particle information. Here the data base format offers fast access(1),
whereas XML storage offers easy data access. KML files are for visualizing
particle tracks in Google Earth. The various ASCII/csv formats can be used
for data exchange with 3rd party tools. If data is written out as individual dfs0
file (one for each particle) the particle class and number is appended to the
specified file name. Additionally, a file with the specified name is generated
that corresponds to the particle count output, i.e. lists the active number of
particles per timestep.

Table 5.3 The different formats for storing particle data

Format Description File type

XML particle track Particle data stored as XML data xml
(compressed/uncom-
pressed)
Particle Track database Particle data stored as binary data- track
base

1 On the file first access a data index will be added/generated, i.e. the file will grow in size, but subsequent data
access will use this index and is much faster.

96 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Output

Table 5.3 The different formats for storing particle data

Format Description File type
KML table particle track Google Earth visualisation kml
TXT table particle track Particle data as simple ASCII/csv txt
data table
Particle count Number of active particles in the txt / dfs0
domain in every timestep, saved as
ASCII/csv or dfs0 file
Track info Particle class information only saved txt
as ASCII/csv, no individual particle
data
Dfs0 Particle track Particle data is saved as individual dfs0
dfs0 file for every particle
Particle track statistics Some general statistics for particle txt
classes and individual particles

NOTE: Outputs generating plain text files should not be used when utilizing
MPI parallelization as the individual sub-domain information cannot automati-
cally be combined. The result file will thus only contain the master domain
data. In this case a standard track (xml or track file type) should be written.
These can be post-processed using the Particle toolbox to convert into the
needed output format.

If a particle hotstart track is generated, i.e. the particle information is intended

to be used as a data file for a file source, only a subset of the formats in
Table 5.3 are suitable. Currently these are the .xml and .track formats.

Particle tracks in XML and binary format can be used within MIKE Zero in the
Plot Composer (Grid Plot/Dfsu Plot), Grid Editor, Result Viewer and Data
Viewer where you can add it as a layer to visualize particle movements (right
click "Add/Remove Layers..."). The KML format cannot be used within MIKE
Zero; its purpose is for displaying data in Google Earth.

Particle track files can get very large, and therefore it is possible only include
a subset of particles and time steps in the output file (see Particle numbers
page 99). Typically, the binary file for a track is smaller than the XML file,
which is smaller than the KML file. Also, the compressed XML file will be
smaller than the uncompressed XML file.

Note that the ‘Particle Track Tool’ in the ‘Oil Spill Particle Track Tools Toolbox’
contains the functionality to convert a particle track in XML format into other
formats such as compressed/uncompressed XML, simple ascii table or dfs0
file format, including the option of filtering the output items.

For each time timestep, the variables for each particle is written in the particle
track file. XML and binary files store the xyz coordinates, mass and additional
variables (see Output Items page 101). KML files store the longitude-latitude

Powering Water Decisions 97

 MIKE ECO LAB / OIL SPILL MODULE

coordinates for viewing the particle track on a map. Further, the z coordinate,
mass and additional variables are also saved and can be displayed in Google
Earth in the Elevation Profile.

XML files come in two variants: compressed and uncompressed.

 In uncompressed format the values of the xyz-coordinates and each

selected variable are saved on separate lines in the file
 In compressed format the values of the xyz-coordinates and each
selected variable are saved in the same line in the file

Figure 5.1 Example of data in XML file using uncompressed particle format

Figure 5.2 Example of data in XML file using compressed particle format

Flood and dry

For 2D field parameters the flood and dry can be treated in three different
ways

 Whole area
 Only wet area
 Only real wet area

Selecting the only wet area option the output file will contain delete values for
land points. The land points are defined as the points where the water depth
is less than a drying depth. When selecting the only real wet area option the
output file will contain delete values for points where the water depth is less
than the wetting depth. The drying depth and the wetting depth are specified
on the Flood and Dry dialog. If flooding and drying is not included both the
flooding depth and the wetting depth are set to zero.

Time step
The temporal range refers to the time steps specified under “Simulation
Period” in the Time dialog.

98 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Output

If checking the "Use simulation end time" check box, then the simulation end
time specified in the Time dialog will automatically be used as last output time
step. Hence, the last output time step will be automatically updated if the sim-
ulation period is changed.

Particle numbers
The range of the particle numbers refer to the particles ID’s in each class.
During the simulation all new particles are given an ID. The first particle num-
ber is 1 and the total number of particles are the multipla of the overall time
step and the number of particles released per time step in each class.

The xml file can get very large, and therefore it is possible not to write particle
variables for all particles, but only a fraction of the total number of particles,
by selecting a frequency higher than 1.

Point series
You must select the map projection (Long/Lat, UTM-32, etc.) in which you
want to specify the horizontal location of the points.

The geographical coordinates of the points are either taken from the dialog or
from a file. The file format is an ascii file with four space separated items for
each point on separate lines. The first two items must be floats (real num-
bers) for the x- and y-coordinate. For 2D field data the third item is unused
(but must be specified). The last item (the remaining of the line) is the name
specification for each point

Line series
You must select the map projection (Long/Lat, UTM-32, etc.) in which you
want to specify the horizontal spatial information.

The line is defined by defining a polyline in the horizontal domain. The

polyline is given by a number of geo-referenced points. A minimum of two
points is required. The polyline is composed of a sequence of line segments.
The line segments are straight lines between two successive points. The geo-
graphical coordinates are taken from the dialog or from a file. The file format
is an ascii file with three space separated items for each of the points on sep-
arate lines. The first two items must be floats (real numbers) for the x- and y-
coordinate (Easting and Northing or Longitude and Latitude). For 2D field
data the third item is unused (but must be specified).

Area series
You must select the map projection (Long/Lat, UTM-32, etc.) in which you
want to specify the horizontal spatial information.

If discrete values is selected for the data type the discrete field data within a
polygon can be selected. The closed region is bounded by a number of line

Powering Water Decisions 99

 MIKE ECO LAB / OIL SPILL MODULE

segments. You must specify the coordinates of the vertex points of the poly-
gon. Two successive points are the endpoints of a line that is a side of the
polygon. The first and final point is joined by a line segment that closes the
polygon. The geographical coordinates of the polygon points are taken from
the dialog or from a file. The file format is an ascii file with three space sepa-
rated items for each of the two points on separate lines. The first two items
must be floats (real numbers) for the x- and y-coordinate. The third item is
unused (but must be specified).

If interpolated values is selected for the data type the values is calculated on
a structured grid (overlay). For each direction you should specify

 The origin
 The grid spacing
 The number of grid points

Furthermore, the orientation of the grid at the origin must be specified. This is
defined as the angle between y-axis in the selected projection and the y-axis
of the grid measured clockwise.

Cross section series

The cross section is defined by defining a polyline in the horizontal domain.
The polyline is given by a number of geo-referenced points which together
make up a polyline. A minimum of two points is required. The polyline is com-
posed of a sequence of line segments. The line segments are straight lines
between two successive points. The geographical coordinates are taken from
the dialog or from a file. The file format is an ascii file with three space sepa-
rated items for each of the two points on separate lines. The first two items
must be floats (real numbers) for the x- and y-coordinate. The third item is
unused (but must be specified). The polyline (cross section) in the numerical
calculations is defined as a section of element faces. The face is included in
the section when the line between the two element centres of the faces
crosses one of the line segments. The faces defining the cross section are
listed in the log-file.

You must also select the map projection (LONG/LAT, UTM-32, etc.) in which
you want to specify the horizontal location of the points.

Domain series
The domain for which mass budget should be calculated is specified as a pol-
ygon in the horizontal domain. The closed region is bounded by a number of
line segments. You must specify the coordinates of the vertex points of the
polygon. Two successive points are the endpoints of a line that is a side of the
polygon. The first and final point is joined by a line segment that closes the
polygon. The geographical coordinates of the polygon points are taken from
the dialog or from a file. The file format is an ascii file with three space-sepa-
rated items for each of the two points on separate lines. The first two items

100 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Output

must be floats (real numbers) for the x- and y-coordinate. The third item is
unused (but must be specified).

You must also select the map projection (LONG/LAT, UTM-32, etc.) in which
you want to specify the horizontal location of the points.

5.21.3 Output Items

Depending on the output type various output items are available. The user
can specify a selection of written output items. For outputs of different type
than "Mass budget" or "Discharge" one can also select among some pre-
defined output specifications of the active output type from the output library.
You can also invoke the output library by pressing the library button.

Field variables
You can select the state variables and the additional variables to be included
in the output file. The list of additional variables is specified in the MIKE ECO
Lab template and can be

 Auxiliary variables
 Processes
 Derived output

If the chosen template defines particle classes it is also possible to select par-
ticle attributes for

 State variables
 Derived variables

The hydraulic variables like velocity components are also included in the list
of additional variables.

You can include/exclude all variables of a category by the control element in

the lower left corner of the dialog. Note that these check boxes are tri-state
elements, i.e. when some elements of a category are included and some
excluded the check box shows an intermediate state.

The "write" column allows you to specify that the current, instantanous value
or some static data (min, max, average in time etc.) will be saved.

In the "handle" column you can specify that the data will be handled normally
or that it will be converted into a mass (total sum), an area concentration
(total sum divided by cell area) or a volume concentration (total mass divided
by cell volume). For particle attributes the default handling depends on the
attribute's type. For mass and concentration attributes the total mass/concen-
tration will be calculated, for other attribute types the average cell value will
be used.

Powering Water Decisions 101

 MIKE ECO LAB / OIL SPILL MODULE

Both statistic values and data handling are only supported for MIKE ECO Lab
variables and particle attributes.

Depending on the output type particle attributes from all suspended, sedi-
mented or particles within a defined z-Range can be selected as own cate-
gory.

Mass budget
You can select the mass budget calculation to be included for the flow and for
the AD state variables (state variables which are transported by the advec-
tion-dispersion). For each selected component the following items are
included in the output file

 Total area - total volume/energy/mass within polygon

 Wet area - volume/energy/mass in the area within polygon for which the
water depth is larger than the drying depth.
 Real wet area - volume/energy/mass in the area within polygon for which
the water depth is larger than the wetting depth
 Dry area - volume/energy/mass in the area within polygon for which the
water depth is less than the drying depth
 Transport - accumulated volume/energy/mass transported over lateral
limits of polygon
 Source - accumulated volume/energy/mass added/removed by sources
within polygon
 Process - accumulated volume/energy/mass added/removed by pro-
cesses within polygon
 Error - accumulated volume/energy/mass error within polygon deter-
mined as the difference between the total mass change and the accumu-
lated mass due to transport, sources and processes

The accumulated volume/mass error contains the contribution due to correc-

tion of the transported component when the values become larger than the
specified maximum value or lower than the specified minimum value.

For temperature the relative energy (relative to the energy at 0 ºC), E (J), in a
volume of water V (m3) is calculated by

E = c water  water T  V (5.20)

where T (ºC) is the temperature, cwater = 4217.0 J/(K.kg) is the specific heat
for water and water = 1010.0 kg/m3 is the density of water.

102 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Output

For salinity the mass of salt, M (kg), in a volume of water V (m3) is calculated
by

M =  water S  V  1000 (5.21)

where S (PSU) is the salinity and water = 1010.0 kg/m3 is the density of water.

Discharge
You can select the discharge calculation to be included for the flow and the
AD state variables (state variables which are transported by the advection-
dispersion). Each selected component will result in a number of output items.

You can select between two types of output items:

 Basic
 Extended

The basic output items are as follows:

 Discharge - volume/energy/mass flux through the cross section

 Acc. discharge - accumulated volume/energy/mass flux through the
cross section

The extended output items that are included in the output file in addition to
the basic output items are as follows:

 Positive discharge
 Accumulated positive discharge
 Negative discharge
 Accumulated negative discharge

By definition, discharge is positive for flow towards left when positioned at the
first point and looking forward along the cross-section line. The transports are
always integrated over the entire water depth.

For temperature the relative energy (relative to the energy at 0 ºC), E (J), in a
volume of water V (m3) is calculated by

E = c water  water T  V (5.22)

where T (ºC) is the temperature, cwater = 4217.0 J/(K.kg) is the specific heat
for water and water = 1010.0 kg/m3 is the density of water.

Powering Water Decisions 103

 MIKE ECO LAB / OIL SPILL MODULE

For salinity the mass of salt, M (kg), in a volume of water V (m3) is calculated
by

M =  water S  V  1000 (5.23)

where S (PSU) is the salinity and water = 1010.0 kg/m3 is the density of water.

Particle track
If your model includes a particle class definition you can select to write out the
attributes of the particles to various track storage formats. Apart from the
state variables and derived variables defined by the particle class you may
also output some generic information such as the age since the release of a
particle and the movement parameters (horizontal direction and speed, verti-
cal speed) used for the latest movement of the particle.

For each selected particle the following items can be included in the output
file

 State variables
 Derive variables
 Age
 Horizontal speed
 Horizontal direction
 Vertical speed

If a particle hotstart track is generated no use selectable output items are

available.

Particle track mass budget

You can select the mass budget calculation to be included for the particles.
For each selected class component the following items are included in the
output file

 Suspended active/current - mass of particles in area within polygon that

is into suspension
 Suspended accumulated - accumulated mass of particles in area within
polygon that has been into suspension
 Sedimented active/current - mass of particles in area within polygon that
is settling on seabed
 Sedimented accumulated - accumulated mass of particles in area within
polygon that has settled on seabed
 Released active/current - mass of particles in area within polygon that is
released within polygon

104 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

Output

 Released accumulated - accumulated mass of particles in area within

polygon that has been released within polygon
 Stranded active/current - mass of particles in area within polygon that is
stranding on dry land
 Stranded accumulated - accumulated mass of particles in area within
polygon that has stranded on dry land

The output library

The output library can be used to maintain different predefined output item
selections, i.e. user defined sets of output items for various output types. A
single selection of output items is called an output set. Such sets can be
stored (and retrieved) from individual files, so called output collections. The
output library consist of at least one or more such collections.

The default collection is the initial values collection and represents the param-
eterisation of the initially loaded setup. It is neither possible to add nor to
remove this default collection or to add/delete sets to it. The main purpose of
the initial collection is be able to revert to the initial states stored in the setup.
To add a new, user defined collection use the "new" button and specify the
storage location and file name in the following dialogue. Please note that
existing files will be overwritten without further notice! The "load" button can
be used to open an existing output collection file, i.e. make all stored output
item selections available in the available definition list (see load output defini-
tion). 'The "unload" button will remove the access to selected collection and
all its sets but will not delete the collection file.

To add the current active output item specifications of the setup, select the
collection and press the "add" button. This will append the collection and add
a new output item set. You can specify a name and a short description text for
both the collection and for each stored set that will be used to identify a spe-
cific set. To update an existing set with the current settings, select the set and
use the "update" button. This will overwrite the stored specifications with the
current values from the setup.

To leave the parameter set library use the "close" or "save and close" buttons.
The later will write any new/updated set (marked with a "*" in front of their
name) to the specified collection files.

Powering Water Decisions 105

 MIKE ECO LAB / OIL SPILL MODULE

106 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

6 LIST OF REFERENCES
/1/ Holmes, D. G. and Connell, S. D. (1989), Solution of the 2D Navier-
Stokes on unstructured adaptive grids, AIAA Pap. 89-1932 in Proc.
AIAA 9th CFD Conference.

/2/ Rodi, W. (1980), Turbulence Models and Their Application in

Hydraulics - A State of the Art Review, Special IAHR Publication.

Powering Water Decisions 107

 LIST OF REFERENCES

108 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S

 INDEX

Powering Water Decisions 109

 Index

A L
About this guide . . . . . . . . . . . . . 9 Line series . . . . . . . . . . . . . . . .99
Aquaculture . . . . . . . . . . . . . . . 13 Living biological organisms . . . . . . . .13
Area series . . . . . . . . . . . . . . . 99 Loadings . . . . . . . . . . . . . . . . .20

B M
Bathymetric data . . . . . . . . . . . . 19 MIKE Animator Plus . . . . . . . . . . .25
Biological processes . . . . . . . . . . 13 MIKE Zero . . . . . . . . . . . . . . . .25
BOD . . . . . . . . . . . . . . . . . . 21 Mussels . . . . . . . . . . . . . . . . .13
Boundary data . . . . . . . . . . . . . 19
N
C Nitrate . . . . . . . . . . . . . . . . . .21
Calibration . . . . . . . . . . . . . 15, 23 Nitrite . . . . . . . . . . . . . . . . . . .21
Calibration data . . . . . . . . . . . . . 20
Check list . . . . . . . . . . . . . . . . 15 P
Chemical processes . . . . . . . . . . 13 Particular matter . . . . . . . . . . . . .13
Phosphate . . . . . . . . . . . . . . . .21
D Plot Composer . . . . . . . . . . . . . .25
Data collection . . . . . . . . . . . 15, 19 Point series . . . . . . . . . . . . . . . .99
Dead or living material . . . . . . . . . 13 Predefined DHI supported ECO Lab
DHI software . . . . . . . . . . . . . . 23 template . . . . . . . . . . . . . . .20, 21
Dissolved substances . . . . . . . . . 13 Predefined DHI supported ECO Lab
DO . . . . . . . . . . . . . . . . . . . 21 templates . . . . . . . . . . . . . . . .13
Project analysis . . . . . . . . 15, 17, 18
E
ECO Lab editor . . . . . . . . . . . . . 22 R
ECO Lab Module . . . . . . . . . . . . . 9 Restricted area search . . . . . . . . . .67
ECO Lab template . . . . . . 9, 16, 21, 61
Ecological modelling . . . . . . . . . . . 9 S
Ecological processes . . . . . . . . . . 13 Seagrasses . . . . . . . . . . . . . . . .13
Ecosystem . . . . . . . . . . . . . . . 22 Sources . . . . . . . . . . . . . . . . .19
Environmental Impact Assessment (EIA) 13 State variables . . . . . . . . . . . . . .13
Eutrophication . . . . . . . . . . . 18, 20 Step-By-Step Training guide . . . . . . .23

F T
Faecal coliform . . . . . . . . . . . . . 21 Tensor . . . . . . . . . . . . . . . . . .95
Flow Model FM . . . . . . . . . . . . . . 9 Total coliform . . . . . . . . . . . . . . .21
Forcings . . . . . . . . . . . . . . . . 19
U
H User background . . . . . . . . . . . . . 9
Heavy metal . . . . . . . . . . . . . . 21
Hypothesis/theory . . . . . . . . . 16, 22 V
Vector . . . . . . . . . . . . . . . . . .95
I
Initial conditions . . . . . . . . . . . . . 20

K
KML file . . . . . . . . . . . . . . . . . 96

110 MIKE ECO Lab / Oil Spill Module FM - © DHI A/S