GRAL Manual GRAL Graphical User Interface 20.01: Amt Der Steiermärkischen Landesregierung
GRAL Manual GRAL Graphical User Interface 20.01: Amt Der Steiermärkischen Landesregierung
ABTEILUNG 15
GRAL Manual
GRAL Graphical User Interface 20.01
Editor
© January 2020
Optional new contour line drawing mode “Spline Line” (chapter 11.4)
Shape line export for contour lines (spline mode only, chapter 11.15)
Support for timeseries of exit temperature and exit velocities for point sources and
portal sources in the transient GRAL mode (chapter 11.11.11)
New options and improvements for wind rose visualization (chapter 14.3)
Copy & paste for objects; copied objects can be rotated in 10 degree steps using
the keys “R” and “L” while the object is floating (chapter 11.12.1)
Delete and add edge points for buildings, area sources and line sources using the
context menu (chapter 11.12.1)
Long taking evaluations got a progress bar showing the real progress and a cancel
button
Some time-consuming functions are started in own threads to reduce the impact to
the UI thread
Vertical profiles: the forms with the vertical graph is refreshed at each new point
instead of creating new forms; this makes it easier to check multiple points
Support for variable stretching factors for the flow field calculation (chapter 8.3)
Object manager: toggle the visibility of a layer using the space key and start the
layout manager for a selected layer using the return key
Item forms: the windows close button and an OK button have been added
Vertical concentration profiles can be visualized in the transient GRAL mode (see
chapt. 15.7.6)
Wet deposition can be considered in the transient GRAL mode (see chapt. 13.2)
The GUI settings can be stored in the user directory (see chapt. 6.1)
Copying the computation cores GRAL and GRAMM to the project folder is optional
(see chapt. 6.1)
More extensive implementation of the local culture for numeric inputs (see chapt.
17.10)
Adapted input dialogs for line sources and tunnel portals (see chapt. 11.11.2)
Double click on items in the “Used source groups” listbox at the sources tab not
mandatory anymore
Support for the new high-performance .NET Core 2.0 GRAL and GRAMM cores
Page 8 of 171 User Guide GRAL V20.01
Changes to previous versions
The German Klug-Manier stability classes in *.akterm files are slightly different
assigned to GRAL classes in order to get a better representation of Obukhov
lengths in the GRAL simulation.
Font size is made adjustable and the automatic scaling can be locked in the wind-
rose chart
Fine tuning to optimize geo-referenced maps has been implemented (see chapt.
11.8)
Bug fixes
The GUI has written invalid decimal separators in the “Topography.txt” file in some
language configurations
Some crashes at the 3D view were fixed
Shape file import and export for area sources produced errors sometimes
In some cases the pie diagram for source group apportionment did not work cor-
rectly
The pollutants SO2, NO2, NH3, and NMVOC have been erroneously taken from
NEMO (= road traffic emission model).
Just another serious error was found and fixed regarding the GRAMM export func-
tion (see chapt. 16.13)
The relative or absolute height and the vertical extension of line sources can addi-
tionally be defined by the user. Please make sure that all source groups to be
used for line sources are defined before creating any line source. Only explicitly
defined source groups are available in the editing modus for line and portal
sources.
New “flat terrain option”, which improves GRAMM flow-field simulations in nearly
flat or completely flat terrain
Additional optional proxy data for buildings introduced: address, house number
New files for the initially seasonal and daily emission modulation introduced:
"Emission_Mod_Diurnal.txt" and "Emission_Mod_Seasonal.txt"
The topography height is displayed on the top right corner of the GIS interface:
when available the GRAL otherwise the GRAMM model cell height.
New option to import the GRAL topography at the GIS window; improved (faster)
import algorithm
Optionally, absolute building and point source heights can be used. Improves
handling of building roofs in complex terrain, i.e. flat roof top instead of terrain-
following roof top.
New object “Wall” has been introduced: Allows for easily digitizing walls, handled
as buildings in the model with a width equal to the horizontal raster size of the mi-
croscale flow-field model.
The editing of source groups for line- and tunnel-portal sources has been facilitat-
ed.
The object manager is actualized, if new items are loaded. Layout managers are
closed automatically, if the redraw order in the object manager is changed, to
avoid wrong assignments between layout and object manager
A serious error was found and fixed regarding the GRAMM export function (see
chapt. 16.13)
Existing projects starting with version 17.08 can be analyzed with the new GUI. Older
projects need to be opened with the GUI version 18.01 or 19.01, the GIS window
(domain window) must be opened and any setting needs to be changed (e.g. the
drawing order in the object manager). This procedure writes setting files in a more
modern format, which can also be read by the current GUI. After this process, the
current GUI can also load older projects.
New projects need to be launched using the actual GRAMM and GRAL versions
20.01. GUI versions up to 20.01 are not compatible with some of the latest output-
formats.
3. Introduction
The following User Guidebook is primarily intended to explain the Graphical User
Interface (GUI) for the Lagrangian Particle Model GRAL (Graz Lagrangian Model)
and the non-hydrostatic mesoscale model GRAMM (Graz Mesoscale Model).
It is explicitly noted that the Government of Styria, Section Air Quality Control, doesn’t
accept liability for software errors and resulting consequences or misuse of the mod-
els. It is the responsibility of the model user to check results for plausibility, which can
be done by comparisons with results from observations other models for instance.
4. Installation of GRAL
GRAL consists of several files and is distributed as a zipped archive. Create any
directory on your computer (application folder), name as desired, and save the
packed archive there. Unzip the files, and best create a link to GRAL_GUI.exe to the
desktop, such that there appears a GRAL-symbol for launching the Graphical User
Interface (GUI).
Microsoft Windows
The GRAL graphical user interface (GUI) is developed for the Microsoft .NET
Framework 4.7 and 64 bit operating systems. An actual framework must be installed
on your computer. It is installed by default in Windows10.
The GRAL GUI creates no registry entries in the Windows system files, so delete the
GRAL folder to erase GRAL from your machine.
Linux
The Linux version of the GUI is developed for the MONO 4.5 framework. Almost all
functions of the windows version are available, except the 3D visualization and the
clipboard functions.
The computation cores GRAL and GRAMM are compiled for the .NET Core frame-
work 3.1, they are platform independent. The windows versions are additionally de-
livered as trimmed single file (standalone package *.exe file) for Windows10. The
.NetCore framework does not have to be installed for this version.
For Linux or advanced windows users, the operating system independent *.dll ver-
sions are available. The .NET Core 3.1 framework is not installed by default. Installa-
tion guides are available at the Microsoft homepage for Windows, Linux and macOS.
The .NET Core 3.1 computation cores are launched from the GUI using the *.exe file
at Windows, the *.dll file at Windows or Linux or - for all supported operating systems
- from the console, using the keyword dotnet (e.g. “dotnet GRAL.dll”).
It is possible to use the Windows GUI, but to start the calculation in Linux.
Due to the different file systems in this case, a small manual adjustment in a control
file must be made if GRAMM wind fields are used. File “windfeld.txt”: add a line with
the Linux path to the used GRAMM windfield data and adopt the path in the file gge-
om.asc (if available).
The files “RecentFiles.txt” and “DefaultPath” are created by the GUI, they are not part
of the delivered zip file.
The folders NETCoreGRAL and NetCoreGRAMM contain the application files and
start files (Windows: *.bat file, Linux: *.sh file) for the .NET Core versions of the com-
putation cores.
We tried to create the GRAL User Interface dpi aware, but it is impossible to test all
possible configurations of screen resolution and zooming options.
If you use a zooming option, it might be necessary to disable the DPI virtualization for
the application. To do this right-click the application’s shortcut and then click “Proper-
ties”. On the Compatibility tab, select “Disable Display Scaling on High DPI Settings”,
and then click OK.
1. GRAL simulation
Dispersion simulation by the lagrangian model GRAL at flat terrain. If you take
buildings into account, a microscale wind filed model is used by GRAL to
compute the flow around obstacles
The GRAL simulation can be performed in a steady-state mode (by default) or
a transient mode.
2. GRAMM computation
Wind field simulation by the mesoscale model GRAMM at complex terrain
GRAMM simulation
Tabulators
GRAL settings Basic settings for the GRAL model (see chapter 8)
Computation Start the GRAL or GRAMM computation and analyze the re-
sults (see chapter 15 and 16.7)
It is important to click on the main project folder and not on any of the subdirecto-
ries when opening an already existing project.
Otherwise, a new project will be created within the subdirectory.
Recently used projects (up to 10 projects) can be opened using the “Open most
recent projects” button.
Add comments to your project by pushing the “Show comments” button. This button
appears as soon a project is opened.
The project-folder name appears at the title bar of the main menu, the complete pro-
ject path is displayed at the bottom of the “Project” tab.
Please avoid any special characters, such as „ä“ „ö“ „ü“ as these may cause prob-
lems.
To be prepared for such cases, there are the following GUI options.
By default, the GUI settings are stored in the directory where the GUI is stored.
Now it is possible, to store the GUI settings in your application data folder (Windows:
My Documents, Linux: home directory). The settings are stored in a new folder called
“GRAL_GUI_Settings”.
This option affects the files for the program settings (default project path, most recent
files) and the default deposition and emission modulation files.
If it is not allowed to launch applications within the project folder through the security
policy on your computer, this setting gives you the option to start the calculation core
in the program folder. The cores will not be copied to the project in this case.
This option needs the GRAL and GRAMM version 19.01 or higher.
As of version 20.01, all item data (buildings, walls, vegetation) of the GUI are written
to the Emissions folder. In the compatibility mode, some data will additionally be
saved to the Computation folder.
7. Save a project
There are two concepts for saving projects at GRAL
All settings and meteorological data at the main menu are saved by pressing the
corresponding buttons at the “Computation” folder of the main form.
If your settings are not already saved and you try to close the project, a warning
message will appear.
At the GIS form all changes are saved immediately, except modifications to the micro
scale terrain (GRAL topography). There is no “save project” or “undo” button at the
GIS form. Your changes at the GIS form are not saved if your project is locked (write
protection - see chapter 15.2).
Modifications of the GRAL topography must be saved with the corresponding menu
entry.
The green lock on the upper right side decides, that the project isn’t locked, that
means no computation results are available and you can edit all parameters (further
information on the lock function can be found in the chapter 15.2.).
Set this value depending on the requested shortest averaging time for concen-
trations, e.g. 1800 s for half-hourly values or 3600 s for hourly values. The lower
threshold of 300 s is due to validity of turbulence parameterizations in GRAL.
Defines the total number of particles released in each dispersion situation. The
total number of particles emitted by the dispersion model is the product of the
parameters “Particles per sec” times “Dispersion time”).
The higher the number of particles, the smoother the concentrations fields (sta-
tistical errors are reduced with increasing number of particles).
Typical values are 25 (for areas < 250x250m²) and 1000 (for areas > 20x20 km²
and numerous sources).
If you need to calculate high stack sources or only a few or single weather situa-
tions, it is recommended to use a significantly higher number of particles.
In most cases (few sources, areas < 1 km²) 100 particles per second are a good
choice.
Note that the calculation time increases linearly with the number of particles.
As of GRAL version 20.01, the GRAL calculation kernel writes the estimated
statistical error at receptor points to the file “Receptor_Timeseries-
_Transient.txt”, which is also generated in steady state mode. The information is
written as soon as the calculation is finished and gives you an indication if a suf-
ficient number of particles have been used.
Surface roughness
If you use flow fields from GRAMM as a meteorological input, and GRAMM was
fed with corresponding land-use data, the local roughness lengths are taken
from the land-use file, which might be inhomogeneous, and the value of this
numeric field is not considered anymore.
Latitude
The number of the dispersion situation to be used to start the GRAL model
(usually the first).
If the “Pause” button in the “Computation” tab is clicked, this input field is updat-
ed with the number of dispersion situation interrupted by the user, such that a
computation can be quickly continued by simply pressing the “Play” button in
the “Computation” tab.
In the “Transient GRAL Mode”, the computation starts with situation 1 or the last
stored intermediate state. Therefore, this setting is not available in the transient
GRAL mode.
With options 1 to 3, all GRAL outputs per dispersion situation (all concentration
*.con, odour *.odr and deposition *.dep files) are stored in one zip container us-
ing the filename extension “*.grz” (gral results zipped).
This option reduces the required disk space by a factor of 2 and the number of
result files in order to the product of source-groups, horizontal slices and calcu-
lation type (odour, with/without deposition).
The creation of the compressed files is incredible fast and there is no influence
on the calculation time.
When compressed result files are used, a compressed ggeom.asc file format is
created. GRAMM and GRAL versions 17.1 or higher are needed to use the op-
tion 1.
The options 2 and 3 have been introduced with GRAL version 20.01 to create
more error-proof files. Option 2 provides smaller result files, option 3 writes all
cells of the grid in linear order.
Buildings
The diagnostic approach is an alternative for very large model domains, if the
prognostic option would lead to very long calculation times and huge memory
consumption.
More information is given some lines further by the description of the column
„GRAL Internal flow field grid“
The prognostic wind field around obstacles is calculated in sub domains, over-
lapping sub domains are merged. Sub domains are defined by the height of the
obstacles multiplied with the default factor of 15. This factor can be overruled to
create larger sub domains. The maximum value for this factor is 1000.
Check this box for carrying out transient dispersion simulations with GRAL. The
plumes of previous dispersion situations are stored in this mode and are con-
sidered again at the dispersion simulation of the next weather situations.
In this way, even puff releases with a minimum release time equal to the user-
defined dispersion time (minimum 300 s) can be modeled. Emissions for each
user-defined source group are modulated in the GRAL calculation core as set
by the user in the file “emissions_timerseries.txt” (see chapter 13.1.2).
to use the option for storing the intermediate GRAL flow field files (.gff files) as
these are then computed only once for each classified dispersion situation.
Note, that simulations times will increase when using GRAL in transient mode
as the entire time series of dispersion situations (as defined in the file
mettimeseries.dat) must be computed one after each other. The standard mode
is therefore the steady-state option (box not checked).
The checkbox “Write vertical concentration file” forces GRAL to write a text file
with all internal stored horizontal mean concentration layers after the last dis-
persion situation is finished.
For computing wet deposition, the GRAL transient mode must be used.
This value defines the horizontal grid size of the concentration grid.
If buildings or obstacles are resolved, grid sizes should be preferably be in the
range of 2 – 3 m.
Whenever the horizontal grid size in GRAL is changed, the model domain – if
already defined – will be deleted for numerical reasons. Thus, the horizontal
grid size must be fixed before the GRAL model domain is defined.
You can define the vertical extension of the concentration grid here. A low verti-
cal extension results in a higher statistical error. Values between one and five
meters would be typical. The concentration is calculated in a layer defined by
the height above ground ± 0.5 times the vertical dimension. The vertical dimen-
sion is also used for the concentration calculation at receptor points (see chap-
ter 11.11.8).
Define the height above ground for each horizontal concentration grid. The
height is defined at the center of the vertical extension. The concentration is de-
termined in a layer of ± 0.5 * vertical dimension.
The lowest height above ground must be higher than 0.5 times the vertical
dimension of concentration layers.
If an odor simulation is planned and you like to use the new concentration-
variance model (see chapter 15.5.2) the lowest height above ground must be
higher than 1.5 times the vertical dimension of the concentration layers, oth-
erwise the new concentration-variance model is not available.
This option defines the horizontal grid size for the microscale flow field.
By default, the grid size of the flow field grid is set to the grid size of the concen-
tration grid. When the grid size for the concentration grid in the “Project” tab is
changed, the flow field grid is automatically adapted. In a few cases it might be
advisable to use different grid sizes for the flow and concentration grids. For ex-
ample, if high stack emissions are still influenced by large buildings/obstacles,
the flow field grid needs to be small, but the concentration grid should take
much larger values to reduce statistical errors.
Define the cell height of the lowest layer of the flow field. Depending on the size
of the model domain, typical values are 1 – 2 m.
Define how cell heights increase with height above ground. A factor of 1.01
means a cell is 1 % higher than the previous lower one.
It should be noted that rather large cell heights may occur in the presence of
mountains, thus it is recommended to use generally stretching factors close
to 1.0 as long as the main memory doesn’t set a limit.
This option enables faster flow field calculations (especially when using terrain)
and/or higher accuracy.
When calculating with terrain, one should be aware that the flow field grid and
thus the vertical cell height starts at the lowest terrain cell within the GRAL do-
main.
This option allows a better accuracy near the ground and faster calculation
times, because there are fewer cells in larger heights above ground. An exam-
ple for wind vectors is shown in the next figure. The default option creates much
more vertical cells, but the vertical resolution near the ground is lower compared
to the option with flexible stretching factors.
If the flexible stretching factors are set, the button color turns to yellow.
The reset button restores the default option without flexible vertical stretching
factors.
Example of vertical wind vectors for default and flexible stretching factors
The number of cells in z-direction defines the number of cells above ground
used in the prognostic flow field model. Above these cells, the wind field is cal-
culated diagnostically.
The total number of cells in the z-direction is automatically defined within the
model (the user cannot change it). The number of prognostic cells in z-direction
in combination with the vertical thickness of the first layer and the vertical
stretching factor determine the relative model height as indicated in the box be-
low. The model height should be high enough that about 90 % of the cross area
in each vertical slice through the model domain is unobstructed.
The option to show the vertical cell heights is for user information only.
Minimum iterations
Define the minimum number of iterations for calculating the prognostic flow
field. Increasing this number can significantly increase the computational times.
Maximum iterations
The model was validated with the default maximum of 500 iterations.
Ticking this checkbox will override the “Maximum iterations” settings and flow
field simulations are running until the internal steady-state condition is reached.
Note that in most cases, simulation times will increase dramatically if you ena-
ble this option.
Define the roughness length of the walls of all obstacles within the model do-
main.
When ticking this box, GRAL writes a file named ‘building_heights.txt’ in the
sub-directory “Computation” if the calculation of the first dispersion situation is
finished. This function is not activated for all other dispersion situations.
Using this option, GRAL writes a file named “GRAL_Topography.txt” that con-
tains the GRAL topography data used in the GRAL model. The topography can
be visualized in the same way as contour maps (for further details see sect.
15.6).
Ticking this box forces the GRAL model to save the calculated flow fields as
files in the subdirectory “Computation” as *.gff files.
This option is particularly useful if more than one simulation must be performed
in the same project (e.g. simulations for several pollutants or for different emis-
sion rates).
This option saves the internal GRAL flow fields around obstacles. When you
create a 2nd simulation, the flow fields are reused to save calculation time. If
*.gff files are saved, it is possible to analyze the data of the flow field (see chap-
ter 17.1).
Using that option saves an additional file for the vertical slices view (file:
“GRAL_geometries.txt“). If you only want to force GRAL to create this file, disa-
ble this checkbox after the calculation of the first dispersion situation is com-
plete and the file “GRAL_geometries.txt“ has been written.
If the model domain, the grid sizes of the flow field calculations, the GRAL
topography or the building geometries are changed (changes are blocked by
the project-lock function for already calculated simulations), all *.gff-files will
be automatically deleted by the GUI as they do not fit anymore to the project
data.
Compression rate
There are two options for the compression rate for “.gff” flow field files:
With the folder button you define a directory for saving and loading the *.gff
files.
If you include *.gff files from another folder, make sure that all GRAL settings
are unchanged. Otherwise GRAL deletes the current *.gff file, calculates a new
flow field and overwrites the existing file.
9. Domain
On the “Domain” tab you can see an overview of the coordinates of the GRAL and
GRAMM domain, the source groups can be defined (see chapter 10) and the GIS
interface can be opened.
Source groups should be defined before sources are digitized and defined. This
allows you to associate sources with source groups.
You can link sources with the same diurnal and seasonal emission modulation to the
same source group. The results of a single source within a source group cannot be
analyzed afterwards.
GRAL calculates concentrations fields for each defined source group (e.g. result A, B
and C in the next figure) and the sum over all source groups to obtain the total con-
centration.
Source groups are needed to set individual diurnal and seasonal emission modula-
tion for the final evaluation of the results (post processing at the steady state GRAL
mode and preprocessing at the transient GRAL mode, see chapter 13). Each source
group can be aligned to one diurnal and seasonal emission modulation.
Edit the name nearby the source group number to add a source group. Unused
source groups are highlighted with a beige color.
You can choose any name for the source-groups, but the comma is reserved and
special characters (not allowed in file names) should not be used.
To remove a source group, delete the name or enter a space for the name. Be care-
ful by removing source groups, because it is not checked programmatically, if
sources are aligned to a deleted source group.
If the GIS interface window is open, any changes made parallel in the main win-
dow, which effect GIS specific features, are not automatically updated in the GIS
interface. For instance, if the horizontal grid size of the GRAL concentration grid is
changed, the model domain will be deleted, this won’t be visualized automatically
in the GIS interface. To do so, the GIS interface must be closed and re-opened
again by pressing the button “Open/Create map” in the tab “Domain”.
All options are available at the menu bar on the top of the GIS window and some
options via shortcuts (take a look to the menu bar or to chapter 19.4).
Depending on the current program-settings and already calculated data, not all op-
tions are available.
Move and hide the toolbox, open the viewframe dialog
Define the GRAMM and GRAL Domain (chap. 16.2 and 11.10),
remove buildings (chap. 11.11.7), 3D view of GRAL and GRAMM
surfaces; generate an animated *.gif file
Object manager / load maps / save actual view / copy to the clip-
board / convert *con files
Load contour maps, mathraster operations, get source apportion-
ment, get concentrations
Edit the emission and basic data for sources, buildings, walls and
receptors (chap. 11.11)
Select or search, filter and sort the respective items (chap. 11.12)
A red lock indicates that results are available, and the project is
protected.
This icon is not available at Linux, but the “GRAL locked” hint is displayed in the title
bar of the GIS form.
Toolbox: symbol
The object manager affects the visibility of the selected items.
The space key toggles the visibility of the selected item. The return key opens the
layout manager for the selected item. The delete key removes the selected item from
the drawing stack.
Toolbox: symbol
Open the Object manager, select the maps or items to be removed and click the
remove button as indicated in the figure below or press the delete key on your key-
board.
There must remain at least one geo-referenced map in the project to set the
coordinate system and to enable correct drawing of all features.
Map or item to be
removed from the
drawing stack
Remove button
You can start the layout manager by double-clicking an item in the object manager
(see chapter 11.3).
Depending on the type of layer listed in the object manager, the layout manager
provides different options for the graphical design.
Select source group The visibility of sources can be filtered and defined by their
source group
Select item Set the line and fill color depending on the parameter selected
here.
In the case of emission sources, pollutants can be selected to
visualize the emission rates in the map by different colors. In
case of line sources also traffic volumes and some more
attributes can be selected.
For buildings, the building height can be selected to represent
the height in defined line and/or fill colors.
Font Change the label font size, color and type using the “Fonts”
Color / Style and “Color” buttons
Line settings Change the line color and width
Color / Width
For the item “Contour lines”: set the contour line color (“Line
properties” – “Color”) to snow white (the default-value) to
show the contour lines in the desired color of the “Fill proper-
ties” – “Line” column. If you set another color at the “Line
properties” – “Color” dialogue, the contour lines are drawn in
that color.
Decimal places Number of decimals places for the labels
Create a color gradient between the first and the last level -
color for line and fill colors
Default Interval Values The computed values when pressing the button “Default” are
classified to the digits 1, 2, 5 and 8 (* 10X).
Equidist. Use this button to get equidistant labels starting from the
lowest level. It is advisable to set the lowest level to the de-
sired value prior to this function.
Default Press the “Default” button to get the default label range
Add Add a new value to the “Values” list box.
Remove Remove the selected value from the “Values” list box.
Save Use the “Save” button to save the color and label settings for
further use
Load Load saved settings for the values and line and fill colors
Fill The drawn item is filled depending on the color in the “Fill” list
box and the transparency option
Show Labels This three-state box offers the following options. If the tick is
set, labels are printed, if the square is selected, labels are
drawn with a frame.
Zoom label font When selected, labels are zoomed and the label size in pixels
corresponds approximately the drawn height in m
Draw border Draws a border around a contour map when the option
“Spline line” is used
Label distance Adjust the label to label distance
Hint: Narrow label distances will reduce the drawing perfor-
mance.
When vectors are selected, the distance sets the interval
between the vectors.
Label interval Set the interval to define, how many contour lines are drawn
between labeled contours. The labeled lines will be drawn
with an increased width, like typically used on topographical
maps.
Min Contour Area If a contour line polygon does not exceed this value, it is not
drawn to avoid “islands”.
Bourke lines Default option; contour lines are drawn using the Bourke
algorithm and filled using rectangles
Spline lines Experimental option; contour lines are drawn using a march-
ing squares algorithm and filled polygons
Filter lines [m] Filter spline line vertices using a Douglas-Peucker filter
Spline tension Set the tension of the used spline (0…1). If the value of the
tension parameter is 0, straight line segments are used to
connect the points
Filter spikes [deg] Remove spikes from the contour line
Vector scale Define the scale of a vector (e.g. wind vectors)
Transparency Set the transparency of filled items using the slider or the
numeric-up down box
Remarks:
By default, there is a tick in the LP-filter box, indicating that the contour map will be
smoothed with a 5-point low-pass filter. This is recommended for most concentration
maps. Uncheck this option when drawing topography maps or other types of maps,
where filtering the data doesn’t make a sense from the point of underlying physics.
Due to compatibility issues, the dimension of point sources and receptors has two
special functions:
GRAL and GRAMM domain show the GRAMM and/or GRAL grid
Vector maps set a smart draw option for a faster redraw of small
Line sources vectors / line sources or buildings
Buildings
Example: how to select a source group or a pollutant for the fill properties
Toolbox: symbol
The number of maps is limited by the memory capacity. The opacity of maps is ad-
justable at the layout manager (see chapter 11.4).
Any map to be geo-referenced must be in first order (on top) in the “Object man-
ager” (to be opened with the button left to the “Open maps” button).
At the geo-referencing process, the maps will be scaled but not turned. So it
is necessary to use north warded maps.
If a pixel graphic is geo-referenced within the GIS interface, the path of the pixel
graphic is stored in the world file. If the project or the pixel graphic is moved to an-
other folder, the GUI will not be able to open this map again but will ask for the
new location of the graphic file. If the graphic file and the corresponding world file
are stored in the same directory (preferably in the subdirectory “Maps” of the
GRAL project), then, GRAL can automatically detect the location even though the
whole project has been moved to a different directory.
The GUI supports geo-referencing using two reference points or one reference point
and a map scale.
During the geo-referencing process navigation of the map is supported either by the
arrow-keys of the keyboard or by the mouse-buttons.
Reference point
Map scale
False values can be removed by clicking the „X“ button to the right of the erroneous
input data. Then click again on the map to correct the data.
During the geo-referencing procedure, the map can be moved and zoomed either
by using the following keys:
+ key: zoom in
or pressing the middle mouse button for moving (see chapter 19.4)
Automatically the top layer in the project is the one taken for the geo-referencing
procedure, thus the map to be geo-referenced must be on top in the list of the “Ob-
ject manager”.
Click on the button as indicated in the figure below. The procedure is the same as
outlined before, though, coordinates of two reference points must be filled in. Click a
second time on the button to close the reference procedure.
World files use file extension like “.pngw”, “tfw”, “.bmpw” or “.jpgw”.
Select the button and adjust the position and scale of the base map using the cursor
keys or the + and – keys.
If the position and scale is set, unselect this menu entry or the button and the new
position is stored.
By pressing the “esc” key the original position is restored.
Automatically the top layer in the project is the one selected for this procedure,
thus the map to be adjusted must be on top in the list of the “Object manager”.
The adjustment of base maps does not affect the corresponding world file.
During this process, the navigation of the map is supported by the mouse-wheel and
the middle mouse button.
Toolbox: symbol
This option is used to save and recall map sections.
Specific map details can be stored by filling in any name in the text field “New view-
frame name” and pressing the “Save” button.
Whenever the map detail is needed, it can be restored by selecting the desired name
from the list box with a double click or by selecting the line and pressing the “Open”
button.
Use the litter button to delete a selected entry from the list.
For a fast access the saved map sections are available in the menu bar of the GIS
window. Simple select an entry to recall the map detail.
Toolbox: symbol
Press the button as indicated in the figure below, then left click on the map to mark
one corner point of the domain. Move the mouse to the opposite corner point of the
domain and release the mouse button to finish the procedure, which can be repeated
at any time. If you press the shift key and the mouse button, it is possible to enter
exact coordinates for the edge points.
It is recommended that no buildings are located nearby the borderlines of the GRAL
model domain, otherwise they should be deleted (see chapter 11.12).
Use the layout manager to define the color and line width of the domain-frame, or to
draw the GRAL concentration raster grid (see chapter 11.4).
It is not possible to define a GRAL domain area, if an invalid GRAMM model domain
exists. In such cases close the GIS window and delete invalid GRAMM information
by clicking the X buttons in the “Topography” tab (see chapter 16.3).
To save all data and finish the editing process, click the box again so that the
checkmark disappears or press the OK button inside the point source dialog.
Data is stored only if a name is given to each source and if the location of
the source has been digitized.
A point source location is digitized by left-clicking the location on the map or you can
enter the coordinates into the textboxes Xpos and Ypos.
Editing window
The drawing options of point sources (diameter, color, show labels) can be set in the
layout manager (see chapter 11.4).
Name identifies the source and a name is needed, to store the data.
Absolute height used to set the point source to an absolute height above sea.
Absolute heights force an error at the GRAL computation core
without topography (see chapter 17.4).
The symbol is used to create a preset for an annual time series for the exit ve-
locity and/or the exit temperature. This time series is used in transient GRAL simula-
tions only.
The button is used to save and reload all settings and the position of the source
at the map is actualized.
Use this dialog to digitize and define the parameters for line sources.
To save all data and finish the editing process, click the box again so that the
checkmark disappears or press the OK button inside the line source dialog.
Data is stored only if each line source is assigned a name, a section num-
ber, a source group and if edge points have been digitized.
Input data is only saved if the „Save SG“ button is pressed at the end!
A line source polyline can be assigned to multiple source groups. This makes sense,
because different pollutants might cause different diurnal or seasonal emission
modulations.
Up to ten different emission rates for different pollutants can be attributed per source
group.
If line sources are digitized manually, the following steps are recommended:
1. Fill in a name, the section, the height above ground or above sea, the vertical
extension and the width
2. Add a source group
3. Digitize the edge point of the line source polygon by clicking to the map
4. Set the pollutant types and the emission rates
5. Press the button “Source groups – Save” (disk symbol) or the store&reload
button (circular arrow)
The pollutant types can only be defined and changed in the topmost item of the
source group list box.
Switch to the
next line source
Use the “Select items” function and the right mouse button (context menu -
see chapter 11.12.1).
Edge point
Name identifies the source and a name is needed, to store the data.
Absolute height used to set the base height to an absolute height above sea
(e.g. for bridges). Absolute heights force an error at the GRAL
computation core without topography (see chapter 17.4).
The drawing options of line sources (linewidth, color) are described in chapter 11.4.
The height of line sources (i.e. lower boundary of the line source) can be defined as
height above ground level (standard value) or optional as height above sea (check-
box “absolute height”). The option absolute height might be useful to digitize bridges.
User Guide GRAL V20.01 Page 61 of 171
GIS Interface
If the absolute height is lower than the GRAL surface, the source is automatically set
to the ground level by the GRAL computation core. If you would like to import line
sources with absolute heights, set the height data at the shapefile to negative abso-
lute values or use the checkbox “absolute height” at the import dialog.
Absolute heights force an error at the GRAL computation core without topography
(see chapter 17.4).
The field “Vert. Ext.” (vertical extension) when editing line sources has different
meanings dependent on its value (because of the compatibility to old projects):
Positive values: the lower boundary of the line source is equal to the ground level
or the value typed in the field “height”, respectively. The vertical extension of the line
source is set to the value provided in the field “vert. ext.”.
Zero: the lower boundary of the line source is equal to the ground level or the value
typed in the field “height”, respectively. The vertical extension of the line source is set
automatically to 3 m.
In order to compute road traffic emissions, the Network Emission Model NEMO 5.0.0
developed by Graz University of Technology must be purchased (Contact: Dr. Martin
Rexeis, Institute for Internal Combustion Engines and Thermodynamics). NEMO
must be stored on the local C-disk as indicated in the next figure. Otherwise, all paths
in the file GRAL.nemo in the working directory of the GUI must be adopted.
A line source can be digitized by left clicking on the map. The actual digitized line
source is indicated during the process. Use a right-click to finish the procedure.
For calculating emissions from road traffic using NEMO, the following input parame-
ters are required: Average daily traffic, share of heavy-duty vehicles, slope, traffic
situation, and the base year.
Page 62 of 171 User Guide GRAL V20.01
GIS Interface
Finish editing and close the GIS Interface. Select the “Sources” tab in the main win-
dow and click the “NEMO” button. You’ll be asked for the location of the NEMO exe-
cutable. Doing this starts the calculations, which can take some seconds up to some
minutes depending on the number of line sources.
By default line source emissions are split into four different source categories:
The user can define the source group number for each of these categories (see also
section 10). Another option is to sum up all sources for each line source, which can
be done by clicking the box (such that the tick disappears) in the window appearing
after the selection of the NEMO executable.
The result of the NEMO computation is not automatically updated if the GIS inter-
face is still open. Close the GIS Interface first and open it again to control NEMO
results.
Start NEMO
To save all data and finish the editing process, click the box again so that the
checkmark disappears or press the OK button inside the area source dialog.
Data is stored only if a name is given to each source and if the location of
each source has been digitized.
An area source can be digitized by left clicking on the map. The actual digitized area
source is indicated during the process. Use a right-click at the second to last point
of the polygon to finish the procedure.
Once an area source has been digitized, the vertices can be edited in a table by
clicking on the button . It is also possible to use the “Select items” function and the
right mouse button (context menu - see chapter 11.12.1).
Each area source is automatically split up into equally sized squares. By default, 26
squares are used to keep the form of the polygon. By changing the raster size, the
number of squares used to raster the area source increases or decreases.
Name identifies the source and a name is needed, to store the data
Absolute height used to set the mean height to an absolute height above sea.
Absolute heights force an error at the GRAL computation core
without topography (see chapter 17.4).
Vert. Ext [m] the vertical extension. Particles are released in a height of
mean height ± Vert.Ext. / 2
The height of area sources can be defined as height above ground level (standard
value) or optional as height above sea (checkbox “Absolute Height”). If the absolute
height is lower than the surface, the source is automatically set to the ground level by
the GRAL computation core. If you would like to import area sources with absolute
heights, set the height data at the shapefile to negative absolute values.
Absolute heights force an error at the GRAL computation core without topography
(see chapter 17.4).
The drawing options of area sources (line width, color, fill options, show labels) are
described in chapter 11.4.
To save all data and finish the editing process, click the box again so that the
checkmark disappears or press the OK button inside the portal source dialog.
Data is stored only if a name is given to each source and if the location of
each source has been digitized.
Imagine looking out of the tunnel: Left-click on the map for digitizing the right corner
of the portal, then right-click to digitize the left corner of the tunnel portal. A symbol
indicates the flow direction out of the portal.
Make sure that all source groups to be used for portal sources are defined before
creating any portal source (source groups are assigned to tunnel portals as with line
sources).
Name identifies the source and a name is needed, to store the data
Absolute height used to set the portal source to an absolute height above sea.
Absolute heights force an error at the GRAL computation core
without topography (see chapter 17.4).
The symbol is used to create a preset for an annual time series for the exit ve-
locity and/or the temperature difference. This time series is used in transient GRAL
simulations only.
An emission computation using NEMO is not foreseen currently, thus the selected
traffic situation is not used at the moment. The selection “bi-directional” or “uni-
directional” has no effect on the dispersion calculations.
The drawing options of portal sources (linewidth, color, show labels) are described in
chapter 11.4.
At gaseous pollutants the deposition dialog allows to edit the deposition velocity. All
other options are greyed and locked.
Computing dust particles (pollutants PM2,5 or PM10) allows the definition of the parti-
cle density. This value and the particle diameter are used by the GRAL computation
core to compute the sedimentation velocity.
In this case it is possible to define fractions for PM10 or PMXX particles. XX stands for
the diameter of the largest particle size. The user can set this parameter.
Pressing the “Reset” button will delete all deposition information, no deposition will be
computed.
The preset is stored in a file called “DepositionSettings.txt”. This file is stored in the
GUI application folder. This text file is tab separated and can be edited and append-
ed using a text editor or a spreadsheet application.
In the case of dust (PM10 or PM2.5) emissions, GRAL calculates the three grain sizes
PM2.5, PM10 und PMXX simultaneously. The largest grain class (PMXX) is only taken
into account for the mass deposited on the ground. If PM10 is calculated, the con-
centration of PM10 in the atmosphere is the sum of the concentrations from the grain
sizes PM2.5 and PM10.
The evaluated “Deposition” files contain the calculated deposition value of all particle
sizes when calculating PM10 or PM2.5. The file name part “Pollutant” “PM10” or
“PM25” solely indicates the calculated pollutant for the dispersion (concentration)
calculation.
Use this dialog to digitize and define the parameters for buildings.
To save all data and finish the editing process, click the box again so that the
checkmark disappears.
Data is stored only if a name has been assigned to the obstacle and if the
edge points have been digitized.
An obstacle can be digitized by left clicking on the map. The actual digitized building
shape is indicated during the process. Use a right-click at the second to last point of
the polygon to finish the procedure.
Use the “Select item” function and the right mouse button (see chapter
11.12.1).
The absolute height is used to set the top of the building to an absolute height above
sea. Absolute heights force an error at the GRAL computation core without topogra-
phy (see chapter 17.4).
The drawing options of buildings (linewidth, color, fill options, show labels) are de-
scribed in chapter 11.4.
Toolbar: button
Buildings and obstacles near the lateral boundaries of the GRAL domain can some-
times cause numerical problems. Use this button to remove these buildings.
Use this dialog to digitize and define the parameters for receptor points.
To save all data and finish the editing process, click the box again so that the
checkmark disappears.
Data is stored only if a name is given to each receptor point and if each re-
ceptor has been digitized.
The minimum height of any receptor needs to be larger than half the height of the
vertical dimension of the concentration layer (see “GRAL settings” tab, chap. 8).
Each receptor can be assigned a value, which can be visualized on the map after-
wards. The following figure shows an example of several receptor points, with differ-
ent values visible by different colors.
The general drawing options of receptor points (diameter, color, labels, display val-
ues) are described in chapter 11.4.
The concentration at the receptor point is determined in GRAL Version 20.01 at the
real position of the receptor with the following volume:
xr Receptor x position
yr Receptor y position
zr Receptor z position
xr ± dx / 2, yr ± dy / 2 and zr ± dz/2
In the Steady State mode, a time series for the receptor concentrations and the me-
teorology at the receptor point can be created using the evaluation buttons in the
Computation tab (see chapter 15.5.1.).
In Transient mode, the GRAL kernel 20.01 or higher writes a time series with the
receptor concentrations and the meteorology at the receptor point. This file is called
“Receptor_Timeseries_Transient.txt”. In addition, the estimated statistical error for
each receptor point is written to this file, also in the Steady State mode, as soon as
the calculation is finished.
Example output for one receptor point and one source group in the Transient mode:
Rec/SourceGroup Rez/SG: 2 Meteo Rez
X 9,4 9,4
Y 43,1 43,1
Z 10 10
11.11.9 Walls
Menu: “Edit - Walls”
Toolbar: checkbox
A wall is a special type of building. The width of a wall is set to the horizontal raster
size of the microscale flow-field model.
The height of each edge point can be set by the user, rising or descending walls are
allowed.
To save all data and finish the editing process, click the box again so that the
checkmark disappears.
Data is stored only if a name is given to each wall and if each wall has been
digitized.
A wall can be digitized by left clicking on the map. The actual digitized wall shape is
indicated during the process. Use a right-click at the second to last point of the
polyline to finish the procedure.
If all edge points of a wall have been defined, the corner points can be edited in a
table by clicking on the button or by using the “Select item” function and the con-
text menu (see chapter 11.12.1).
The option absolute height is used to set the top of the wall edge points to absolute
heights above sea. Absolute heights force an error at the GRAL computation core
without topography (see chapter 17.4). A mixture between absolute and relative
heights along one wall is not possible.
11.11.10 Vegetation
Menu: “Edit - vegetation”
Toolbar: checkbox
Use this dialog to digitize and define the parameters for vegetation areas.
To save all data and finish the editing process, click the box again so that the
checkmark disappears.
Data is stored only if a name is given to each vegetation layer and if each
vegetation layer has been digitized.
Vegetation can be digitized by left clicking on the map. The actual digitized vegeta-
tion shape is indicated during the process. Use a right-click at the second to last
point of the polygon to finish the procedure.
The following properties can be set for vegetation (note that there is a drop down
menu with some default values for different types of vegetation to facilitate selecting
proper input values):
Trunk zone [%]: height of trunks in percentage of the total height, e.g. if the total
height is 10 m, and the trunk zone is set to 20 %, then the trunk
is assigned a height of 2 m.
Leave area densi- values separated for the trunk and the crown zones.
ties [m²/m³]
Coverage [%]: defines the area occupied of the specified vegetation type with-
in the digitized area in percentage of the total area.
Once vegetation has been digitized, the vertices can be edited by clicking on the
button or by using the “Select item” function and the context menu (see chapter
11.12.1).
The drawing options of vegetation (line width, color, fill options, show labels) are
described in chapter 11.4.
Therefore, there is the button in the point source and portal source dialog.
To create a new entry, simply add a new name and press the button “Create new
entry”.
If you like to assign an existing entry to the source, use the list box “Select an entry”
and the existing values are displayed.
The button “Exit and save” writes the values to a file to the project folder “Emissions”
and assigns the time series to the source.
The button “Restore” removes the time series from the recently edited source.
When you create the emission files (tab computation in the main window) and time
series are assigned to sources, time series files are written inside the folder “Compu-
tation”. The files are called
TimeSeriesPointSourceVel.txt
TimeSeriesPointSourceTemp.txt
TimeSeriesPortalSourceVel.txt
TimeSeriesPortalSourceTemp.txt
These files are simple text files. When needed, it is possible to edit the values for
each hour of the time series within these files.
Toolbox:
Click on the corresponding selection button or menu entry in the GIS Interface to
select a desired item.
If you select an item with the left mouse button, this item will be highlighted, and an
info-box will open with relevant information about the selected source.
A selected item can be edited quickly by setting the tick in the editing box, or it can
be easily deleted by pressing the “delete” key.
If more items of the same kind (sources, buildings) are selected, the complete select-
ed group can be deleted by simultaneously pressing the “shift” and “delete” keys (see
chapter 19.4) or the corresponding menu item “Edit – Delete selected items”.
Select an item with the right mouse button to open a context menu (see chapter
11.12.1).
Move move the item, replace it using the left mouse key
Copy create a copy, replace the copy using the right mouse key
and use the context menu entry “Paste”
During copying, objects can be rotated left or right in 10 degree steps using the R
and L keys while the object is floating.
Toolbox:
Use the menu entry “Select –Search items” or the “magnifier glass” button at the
“Select items” section of the toolbox to open this menu.
This menu is designed to display all items (sources, buildings, receptor points) in a
data grid. It is possible to sort and to filter the items, as well as copy the information
and to select an item.
You can sort the items by clicking to the top of the data column.
There are filters for the name, the type, the height and the source group (SG).
The filter for the height is a decimal filter, therefore you must use logical operators,
like “=”, “<”, “>”, “<=” and “>”.
Example: Point sources (Po), height >10m, all source groups, all names
If you select an item (use the left column with the triangle to select a line) and double
click the triangle or the ”OK” button, the map will display the desired item and open
the corresponding “Edit item” dialog.
If sources of a type appear with the same “Nr.”, take a look to the column
“Source_group”, because this source will be assigned to multiple source groups.
You can use a Right click in the header of the data table to hide or show additional
columns.
Toolbox:
Shape files (.shp) containing information about sources, reference points, buildings
or walls can be imported using the buttons in the GIS Interface or the menu entry.
Tunnel portals cannot be imported from shape files but from other GRAL projects.
Meteorological data can be imported to visualize wind roses within the GIS window.
Importing sources, reference points, or buildings from already existing GRAL projects
is quite easy: After clicking the corresponding import button in the GIS Interface se-
lect the path and file to import. Note that the GUI stores all sources in the subdirecto-
ry “Emissions”, while buildings and receptor points are stored in the subdirectory
“Computation”.
Subsequently the import of line sources is described briefly. Importing other features
is done in a similar way.
After clicking on the corresponding button to import line sources, an additional win-
dow is opened, where the attributes of the shape file (stored in a .dbf file) can be
selected.
Such can be for instance building heights, though only uniform heights can be de-
fined for all buildings to be imported. Individual building heights can be manually
adopted in the list view.
Page 82 of 171 User Guide GRAL V20.01
GIS Interface
Emission rates are required in [kg/h]. In case that those are only available e.g. in
[kg/a], a common conversion factor can be applied when importing sources (in this
example to convert from [kg/h] to [kg/a] the conversion factor would be 0.00011.
Alternatively it is possible to check the checkbox ^(-1) and use the conversion de-
nominator, in the example 8760).
Only red indicated attributes must be defined in any case when importing features.
All other (marked blue) attributes are thus optional.
Add columns. Simple
Define the depositi- mathematical operations
on settings with any other existing
column are possible to
derive new values for an
attribute.
If you define deposition settings (see chapter 11.11.5), they are assigned to all
imported sources.
If the absolute height checkbos is checked, all imported heights are interpreted as
absolute heights above sea. Absolute heights force an error at the GRAL computa-
tion core without topography (see chapter 17.4).
User Guide GRAL V20.01 Page 83 of 171
GIS Interface
1 Point
3 PolyLine
5 Polygon
11 PointZ
13 PolyLineZ
15 PolygonZ
The *.dbf Reader was developed by Ahmed Lacevic and is licensed with the BSD 2
Clause “Simplified” license.
Toolbox:
It is possible to visualize wind roses within the GIS window. The imported *.met files
need an information about the measurement coordinates. This is an extension to the
*.met file (compatible to the GUI 19.01 and older).
It is needed to add the coordinates and possible to add further information using the
characters “//” without any separator characters in the meteo file “*.met” header, e.g.
//X=80079
//Y=233949
//Z=10
01.01.2013,00:00, 2.1,218,6
01.01.2013,01:00, 3.2,204,4
……….
New wind statistics (from GRAMM wind fields) are created with this header automati-
cally.
Size [m] Diameter of the wind rose in m (the wind roses are zoomed)
Example:
Toolbox:
All emission sources (except for tunnel portals), receptor points, or buildings can be
exported as shape file, and in this way can be used in other software tools or subse-
quent GRAL projects.
Simply click the corresponding export button for the feature to be exported and define
a file name and directory.
Not all attributes will be exported, but only those that are required for further eval-
uations or dispersion calculations.
The topmost contour map within the object manager (see chapter 11.3) is exported.
The export of contour lines is limited to contour lines with drawing option “Spline
Line” (see chapter 11.4).
If the GRAL domain area or the GRAMM domain area is defined, the domain borders
and the numbers of concentration and flow field cells (if buildings are considered) are
displayed at the “Domain” tab.
Number of cells
Number of cells
Please note that high cell counts increase the calculation time. If the calculation time
is very high, it is recommended to reduce number of cells by increasing the horizontal
grid resolution or setting smaller domain areas. Note that a larger horizontal flow field
resolution may result in a poor building and surface resolution.
After the editing process of all sources is completed and the GRAL model domain
has been defined, close the GIS Interface. Select the “Sources” tab in the main
menu, in which all defined source groups can be selected for subsequent dispersion
calculations.
Mark source
groups
Remove a selected
source group
Modulation preview
Select the desired source group entries in the list box “Available source groups”.
Multiple selections are possible using the shift or ctrl button. Use the arrow button to
move the selected items to the list box “Used source groups”.
A simple click on an entry in the list of used source groups updates the modulation
preview.
Double-clicking on a selected source group item opens the edit form for the emission
modulations.
It is possible to set monthly and hourly emission variations. There are some pre-
defined modulation functions, although custom modulations can also be added.
By pressing the „Show emissions“ button, annual totals for the selected pollutant are
calculated and displayed in a diagram.
In the lower left corner of the diagram a click box allows for calculating annual sums
with or without the defined emission modulations (assuming a constant emission
modulation of 1 for each hour of the year). In this way, the effect of the defined emis-
sion modulations on the average emission rate for one year can be checked quickly.
Select pre-defined
seasonal variation
To add user defined seasonal and diurnal emissions, you must click to the corre-
sponding button, define a name and edit the values in the data grid. Only numerous
values are accepted, you must use the standard decimal separator.
If a new modulation name already exits, the character “_” is automatically appended
to the new name.
The values are temporarily stored and visualized if you press the “Save” button. The
values are updated in the model when you confirm all settings with the “OK” button.
You paste values from a spreadsheet application by clicking in the data grid view
(selecting the grid, do not switching to the edit mode) and pressing the “Control - V”
keys on the keyboard.
It is possible to change the modulation for each source group after the computation is
complete, since the evaluation of the results uses the current modulation data (see
the flow charts in chapter 5). This is an easy and fast way to evaluate different emis-
sion modulations after a computation has been finished.
Presets of the emission modulation are stored in the application folder in the files
"Emission_Mod_Diurnal.txt" and "Emission_Mod_Seasonal.txt". The data inside
these files is tab separated and can easily be edited by any text editor or spread-
sheet application. It is possible to edit or add own presets in these files. For more
information about the file formats look to chapter 19.3.
When utilizing one of the routines „Multiple Sources“, „Mean, Max, Daily Max“,
„Percentiles“ or “Receptor Concentrations” emissions modulations for every
source group can be defined for each hour of a year. When doing so, the pre-defined
emissions modulations for monthly and daily variations are not considered.
Hourly emissions modulation factors for each source group must be stored in a sepa-
rate file called “emissions_timeseries.txt”, which needs to be stored in the sub-
directory “Computation” of the current project.
This file can be created by clicking the button “Emission time series - Create” in
the emission tab. This function uses the preset values of the diurnal and sea-
sonal emission modulation and writes the file “emissions_timeseries.txt” to the
computation folder.
The generated file can also be visualized, using the button “Emission time se-
ries – Show”.
It is imperative to use the date and time information as stored in the file
“mettimeseries.dat”, which can be found in the sub-directory “Computation” of the current
project. It is recommended to copy the file in an application such as Excel. In a next step
all columns, except the first two ones containing the date and time information, must be
deleted.
The first column is the date, the second the hour, followed by the numbers of each used
source group. It is not important in which order the source groups are aligned.
For more information about the file formats look to chapter 19.3.
In the 1st step select the pollutant and select the checkbox “Enabled” in the frame
“Wet deposition”.
These settings are a property of the pollutant and apply to all source groups.
The washout rate is computed in analogy to VDI 3945/Part 3. If the wet deposition is
enabled, a file Precipiation.txt will be written into the computation folder. This file
contains precipitation data for the model domain for each entry of the time series. By
default, all values are set to a precipitation of 0.0. You can replace these values by
given precipitation data in mm/h.
User-defined decay rates λ in s-1 reduce the particle mass every time step by apply-
ing an exponential function:
𝑚𝑡+1 = 𝑚𝑡 𝑒 −𝜆𝑡
Decay rates might be used to simulate inactivation rates of e.g. bacteria or radioac-
tive decay.
The decay rate is a property of the pollutant and applies to the defined source
groups. A custom decay rate per source group can be defined.
Click the „Open Met-File“ button to import meteorological data in the „Meteorology“
tab of the main window. Two different input formats are provided:
5th column: stability class – 1=strong convective (A) to 7=strong stable (G)
For best support of all features, it is recommended to add a 3-line header with the
coordinates of the measurement station (X, Y: coordinate, Z: relative height above
ground).
Example file
//X=116093
//Y=255564
//Z=10
01.08.2017,00:00,0.7,227,7
01.08.2017,01:00,1,210,7
01.08.2017,02:00,1.5,217,7
01.08.2017,03:00,1.1,215,7
Select file
Make sure meteorological data has been imported correctly by selecting the cor-
rect separators for the decimal places and columns. Press the „Check“ button to
visualize how the meteorological data is aligned.
Press the „OK“ button if the data preview is displayed correctly and the calendar
functions are displayed at the bottom.
Preview met-data
Use the calendar to select a time span for the dispersion simulations.
If you want to choose a different meteorological time series, just repeat the proce-
dure.
Data can be analyzed by using the buttons on the left side of the window.
Press the „Classification/Default“ button to define the number of wind speed and wind
direction sectors. The standard wind speed bins are designed for regions with fre-
quent low wind speed situations. These default values can be reloaded at any time
by pressing the „Classification/Default“ button again.
If you check the “No classification” box, no classification will be made. Instead, every
hour (half-hour) of the loaded time series is calculated.
Don’t forget to set or check the anemometer height after importing a *.met file!
Analysing tools
The meteorological data import function can be used to analyze such data at any
time, even after a user has already selected a specific meteorological file for use in
the dispersion simulations.
A new meteorological input for the dispersion calculation will only be created, if
you press the „Classification/Default“ button.
It is possible to analyze other meteorological files, but the prior selected meteoro-
logical file for the dispersion calculations is still used.
If you are not sure which meteorological file is used for the simulations, just close
the project and open it again or look at the bottom line of the “Computation” tab –
“Meteorological input”.
Windrose Windrose/Stability
The bias correction is useful for binned meteorological files, e.g. for an artificial
equally distributed wind rose.
Without bias correction Bias correction applied
Select the “Computation” tab and press all buttons to the right with a red circle.
The “Meteo files” button generates GRAL specific input files from the selected mete-
orological input file.
Source files and building data in GRAL specific formats are generated by pressing
the corresponding buttons. However, this data defined with the GIS Interface are
stored anyway and aren’t lost, if the buttons are not pressed before the project is
closed.
You can also define the number of processor cores to be used in the simulations. By
default, the maximum available processor cores of the computer in use are suggest-
ed.
After each button has been pressed, a new group box appears at the top right of the
window. The simulation can be started by pressing the right button (“Play”). In the
next step the user is asked for a GRAL executable.
When you start a simulation for the first time, the GUI suggests an executable from
the parent directory, where the GUI and all related files are stored. Subsequently,
whenever GRAL is started again, the GUI suggests the executable, which has been
copied from the parent directory of the GUI to the current GRAL project in the “Com-
putation” subdirectory. In this way, it is possible to re-start the simulation with the
same executable even after years.
If you want to run a simulation again, but with a newer version of GRAL, either man-
ually copy the newer version to the “Computation” subdirectory of the current GRAL
project or delete the old executable file before starting GRAL with the GUI.
A user can press the “Pause” button at any time, which immediately interrupts the
simulation and saves the current number of dispersion situation in the corresponding
input field in the “Project” tab. If you press the “Play” button again, the simulation will
start with the previously calculated dispersion situation.
Stop GRAL
Parameters for the post-processing, like daily and diurnal modulations, can be
changed in a locked project (see chapter 13.1).
If you need to restart a project simulation at a higher dispersion situation, you can do
this in a locked project. Simply set the situation on the “GRAL settings” tab – “Start
with dispersion situation” - and switch to the “Computation” tab.
Now you must unlock the project manually by pressing the lock symbol and
confirm the warning message.
In the next step you can click on the “Control files” button and to restart the calcula-
tion at the selected situation.
Typically, GRAMM simulations are performed with a rather coarse horizontal resolu-
tion, e.g. a few hundred meters. In some applications it might be desirable to dissolve
the terrain as fine as possible.
To do so, activate the check box in the ‘Computation’ tab as shown in the next figure
or use the menu item “File – Import – Original GRAL topography” in the GIS window
(see chapter 17.6). When you use the menu entry, the GRAL cell height is displayed
in the GIS-window and you can edit and visualize the GRAL topography (chapter
17.7).
If you select the check box, you must select a (high resolution) terrain file at the start
of the GRAL computation. The GUI automatically generates a subset of this terrain
file with the exact dimensions and resolution as defined for the GRAL simulations.
This new terrain file will then be used by GRAL as lower boundary.
If you restart GRAL, e.g. to calculate the dispersion for another pollutant, the already
existing topography file will be used. Any change to the GRAL domain or resolution
will delete the high-resolution topography file and it must be generated again to
match with new dimensions and/or resolution.
Toolbar:
If your simulation considers buildings and/or walls, it is recommended to check the
“Write file building_heights.txt” checkbox (see chapter 8).
If this checkbox has been checked, it is possible, to visualize the original rasterized
GRAL buildings and walls at the GIS interface.
To do this, the GRAL computation must be started. After the 1st meteorological situa-
tion has been finished, the file “buildings_heights.txt” will be written. At this moment
break the computation and open the GIS interface (chapter 11).
Load buildings_heights.txt
from the folder computation
Just load the file “buildings_heights.txt” from the computation folder, like a contour
map. The buildings are drawn rasterized as they are used in the model. The color
depends on the building height and can be defined at the “Layout manager” (see
chapter 11.4).
There are several methods to analyze the results after the computation has been
finished.
Calculate the average concentrations for the imported meteorological time se-
ries. Concentrations are calculated for each source group and additionally for
the sum for all source groups. Note that this function is much faster than the
one below. This function uses the diurnal and seasonal modulation for each
source group, the “emissions_timeseries.txt” is not used by this function.
„Receptor Concentrations “
In case of existing receptor points (these must be defined by the user prior to
any simulation) this procedure can be used to get time series of concentrations
for each receptor point and source group. The result is stored in the file “recep-
tor_timeseries.txt” in the subdirectory “Computation” of the GRAL project. This
Page 106 of 171 User Guide GRAL V20.01
Starting GRAL and analyzing results
file can be further processed with any suitable software, e.g. Excel. This func-
tion uses the diurnal and seasonal or the “emissions_timeseries.txt” file modula-
tion for each source group.
In addition, wind speed and –direction as simulated with the microscale flow
field model of GRAL are exported to files named ‘GRAL_*name*.met’ and
stored in the subdirectory ‘Metfiles’, in case that buildings have been considered
in the GRAL simulations. These files can be read and visualized as outlined in
chapter 14.
In GRAL transient mode, the GRAL computation kernel writes a file called “Re-
ceptor_Timeseries_Transient.txt” with a time series of concentrations for each
receptor point and source group. You do not need to start this evaluation in this
case.
„Percentiles“
Compute the user defined percentile. Percentiles can be selected from the in-
terval between 90 and 100. The function requires a lot of memory, so it might
happen that an error message occurs when the available memory on the com-
puter is exceeded.
This function uses the diurnal and seasonal or the “emissions_timeseries.txt”
file modulation for each source group.
The results of the above described procedures are stored (except those for
“Receptor Concentrations”) in the subdirectory “Maps” of the GRAL project.
The results are stored as modified Esri ASCII Files (see section 19.3.1 for more de-
tails), using automatically generated file names. You can specify an individual file –
prefix at the “Select Source Groups” dialog. This prefix can be used to separate the
results of several evaluations (e.g. different diurnal or seasonal emission modulations
- chapter 13.1). Use the GIS Interface to visualize the results files as contour maps.
Examples
Mean_NOx_Linesource_8m.txt
98_Bioaersols_total_3m.txt
Deposition_Mean_NOx_total.txt
Deposition, mean value, pollutant “NOx”, total = sum of all source groups
The evaluated “Deposition” files contain the calculated deposition value of all particle
sizes when calculating PM10 or PM2,5. The file name part “Pollutant” “PM10” or
“PM25” indicates the calculated pollutant for the dispersion.
Compute odour hours in [%] with regards to the meteorological time series
used. Odour hours are computed for each defined source group as well as sum
for all source groups.
Typically, the sum for all source groups gives not the same figure as
when simply adding the figures at a receptor point for non-linear rela-
tionships resulting from the definition of an odour hour.
„Compost“
This procedure has been developed to compute odour hours in [%] in the vicini-
ty of compost works. Up to three different processes, each owing certain emis-
sion rate [MOU/h] and frequency for a year [%] can be defined by the user.
Compute odour hours [%] in the vicinity of animal farms applying the so-called
all-in-all-out system, where all breeding animals are getting in the stable and fi-
nally put out of the stable together at once. The function considers linearly in-
creasing odour emissions for a breeding cycle. The user will be asked for the
number of days of an average breeding cycle.
In each of the procedures above the user is asked for an odour threshold (e.g.
1 OU/m³, 3 OU/m³) and the ratio of the 90 percentiles of the odour cumulative fre-
quency distribution and the average odour concentration for an hour. By default, a
factor of 4 is suggested by the GUI.
Since GRAL version 17.1 and newer, as an alternative a more elaborate model that
computes the factor in a more realistic way is provided (concentration-variance mod-
el). It is generally recommended to use this model for assessing odour hours.
This option is available, if the lowest height above ground for the concentration
grid is higher than 1.5 * vertical dimension of concentration layers (see chapt.
8.2.).
Toolbar:
Concentration maps and other kinds of contour maps (e.g. topography maps) can be
visualized in the GIS Interface. Click the button as shown in the next figure or use the
menu entry “Analyze results – Load contour map” and select the desired file, which
needs to be in ESRI-ASCII format (see section 19.3 for more details).
A variety of options are offered to design such contour maps. Just open the layout
manager (see chapter 11.3 and 11.4) for the desired contour map.
Note that it can take rather long to draw large contour maps, while it is much faster to
draw pixel maps. To achieve this, the line width must be set to zero. The GUI sets the
line width initially to zero if the number of cells exceed a value of 0.2 million.
Toolbar:
GRAL stores concentration files for each dispersion situation in the subdirectory
“Computation” of the GRAL project as *.con (concentration), *.dep (deposition), *.odr
(odour) or the compressed *.grz file format.
Such a file for a single dispersion situation can be converted in a readable ASCII
format by pressing the “.con .txt” button or the “Export – Convert *.con files to ASCII
raster format” menu. The converted files are stored in the subdirectory “Maps” of the
GRAL project and can be visualized immediately like any other contour map.
Toolbar:
All maps stored in the ESRI-ASCII format (see section 19.3) can be further pro-
cessed by applying user defined mathematical expressions. Up to two maps can be
processed together (e.g. computing the difference). A simple empirical relationship to
convert NOx maps to NO2 maps is provided.
Constants pi e
Note: The mathematical parser “Mathos Project” has been developed by Artem Los.
This part of the GRAL code underlies the BSD-3 license.
The unit of the result is taken from the input raster file A, but can be changed manu-
ally by the user (e.g. conversion from µg/m³ to mg/m³ or mg/m².d to g/ha.Y).
Toolbar:
Calculation and visualizing of the source group assignment (pancake diagram) at any
location. This option is useful if more than one source group was defined before
simulations were started (chapter 15.7.3)
Click on the map to define the position of your analysis and select a
“Mean*_total*.txt” file.
The function reads the concentration at the selected position from all available
source group maps and displays the result.
The source group maps are searched by the selected file name, but the string part
“total” is replaced by an asterisk “*” character. If you use this function, you should
ensure that there are no “old” or invalid maps in the map folder.
Note that the procedure asks for the background concentration, which is added if
desired.
Toolbar:
Extracting values of any map at a certain position (chapter 15.7.4)
Using this function, you can extract computed concentration values for specific points
to get a quick info of the simulated concentration at one raster point without low-pass
filtering.
In a 1st step you can select the desired concentration file or dispersion situation. In a
2nd step, simply click on the GIS form to get the concentration values at the selected
points.
When selecting this function, a new entry can be found in the object manager, called
“Concentration values”. Use this entry to show and hide the values or to define the
size of the symbol or the font.
Toolbar:
In the first step, you need to convert and load a single *.con file using the function
from chapter 15.7.1.
In the second step set the visualization options using the layout manager (chapter
11.4).
After these preparations start the GIF recorder using the button above.
Set the 1st and final situations and the time per frame. Using the Save-Button you can
define a folder and file name for the GIF file.
The press Start and the GUI loads and converts the *.grz files, displays the *.con file,
makes a snapshot and stores the picture to the GIF file. The recording button turns to
. You can stop the recording, using this button.
Toolbar:
User Guide GRAL V20.01 Page 115 of 171
Starting GRAL and analyzing results
If the GRAL simulation has been performed in a transient mode and all dispersion
situations within the “mettimeseries.dat” file are computed, it is possible to show
vertical profiles at specific points.
Simply click on the GIS form to get the concentration profiles at the selected points.
The z-axis can be zoomed, using the + and – button in the vertical concentration
form.
The profile is refreshed at each new point if you select a new position.
Here are examples of vertical profiles for a single dispersion situation (stable, low
wind speed) along the plume of a stack source
Toolbar:
If the GRAL simulation has been performed in a transient mode and all dispersion
situations within the “mettimeseries.dat” file is computed, it is possible to show verti-
cal profiles along a section line.
In the next step you can set a filename and the maximum height of the vertical pro-
file.
This function creates an ESRI ASCII File along this section line with the given hori-
zontal resolution of the concentration grid. The vertical resolution is set to the same
value, the GRAL internal resolution will be much higher.
ESRI ASCII files can be visualized by the GUI and this file will be visualized as a
concentration grid, except the additional x and y axis and the white background. All
settings for concentration maps can be used (see chapter 15.6).
Here is an example of the vertical concentration along a section line for a single dis-
persion situation (stable, low wind speed) and the plume of a stack source (the pic-
ture is stretched in vertical direction).
Due to the interpolation for different raster widths, the concentration maps of the
vertical section and horizontal section may differ. The vertical profile is much more
accurate than the vertical section.
The user must define the horizontal grid resolution, the height of the lowest cell (in
most cases 10 m is useful), the number of cells in the vertical direction, and the verti-
cal stretching factor.
The stretching factor must be chosen in a way, such that the top of the model domain
is at three times as high as the highest elevation in the domain. The top of the model
domain is indicated in the field “Relative top-level height”, which is the relative height
from the lowest level in the domain.
The model domain itself should be defined after having fixed these parameters.
Every time the horizontal resolution of the GRAMM grid is changed, the
GRAMM domain is deleted for numerical reasons.
Define GRAMM
grid
Toolbar:
Start the GIS Interface (“Domain” tab) and click the button as indicated in the next
figure or use the menu entry “Edit – GRAMM model domain”. The left mouse (left
mouse button pressed) is used to drag a rectangle. This procedure can be repeated
several times.
Clicking this button allows the user to first set the number of cells at the model
boundaries for which the topography is smoothed. The number of cells multiplied with
the chosen horizontal grid spacing indicates the spatial extent of this zone. The ex-
tension of the smoothed zone at the boundaries should be defined in dependence on
the relative height between the highest and lowest elevation within the model do-
main. The higher the relative elevations within the model the larger should be the
smoothed zone. It may vary between a few hundreds of meters and a few kilometers.
Afterwards the user can select the topographical data, which must be stored in ESRI-
ASCII format (see section 19.3).
If meteorological data has already been imported, GRAMM simulations are ready to
be launched at this step.
The GUI generates automatically the file “ggeom.txt” in the subdirectory “Maps”,
which can be used to visualize the model topography. It is recommended to generate
a contour map prior to any GRAMM simulations to check if the topography has been
generated in a right way. As an option use the 3D view (see chapter 16.6).
Use the “X” button to delete all GRAMM information (GRAMM area, GRAMM set-
tings).
In case that no input file for land-use data is available, at least a homogenous land-
use class can be defined for the model area. Important: The code for the default
land-use category is 0 (zero). By pressing the button as illustrated in the following
Page 120 of 171 User Guide GRAL V20.01
Flow field simulations with GRAMM
figure, a table is opened, where the characteristics of the default land-use category
can be defined.
In addition, own land-use categories can be defined in this table. Note: The codes
are limited to the range 0 – 999. In this way, the soil characteristics of own land-use
categories with individual codes can be defined. These will be applied when import-
ing a corresponding input file containing these codes.
Basically, the CORINE classes (codes with 3 digits) are implemented. These pre-
defined CORINE classes can also be changed, by providing new soil parameters in
the table for each class.
All redefined land-use classes are stored in the file “Landuse_Default.txt”, which is
saved in the sub-directory “Computation” of the project. The file can also be edited by
using simple text-editing software.
The format for all files containing land-use data is ESRI-ASCII (see section 19.3.2).
Such files can be loaded by pressing the „Create GRAMM land use file“ button as
depicted in one of the figures below.
The horizontal resolution of this landuse file needs to be less than or equal
to the horizontal grid size of the GRAMM grid used.
New land-use
categories
In case that land-use data has been imported, the following files are automatically
generated in the subdirectory “Maps”, which can be used to visualize surface param-
eters and check for validity:
Since GRAMM 17.01 an improved time step adjustment is used in GRAMM making
the choice of the maximum allowed time step less critical. Now it is possible to use a
maximum time step of 10 to 30 s regardless of the horizontal grid size.
Up to GRAMM version 17.01 the following values for the maximum time step and
relaxation factors were recommended to ensure numerical stability of GRAMM simu-
lations:
All control parameters filled in for GRAMM simulations are stored automatically.
There is no need for extra confirmation. Topographical data as used in GRAL simula-
tions can be generated by ticking the checkbox “Write file ‘GRAL_Topography.txt’”.
The corresponding file is stored during the computation of the first flow field situation
in the sub-directory ‘Computation’. It can be viewed as any other contour map.
“Sunrise option”
If this option is checked, GRAMM will use an alternative approach to compute the
flow fields using modelling times up to 6 hours to simulate sunrise or sunset effects.
The calculation time and the number of resulting flow fields increases, because in-
termediate results are stored, too. This option is useful for computing large valley-
wind systems by making use of artificial dispersion situations for the GRAMM initiali-
zation in combination with the match-to-observation function after finishing the simu-
lations. GRAMM numbers automatically these intermediate flow fields and stores
them correspondingly in separate files. To make these intermediate flow fields avail-
able for the match-to-observation function, additional “dispersion situations” are in-
troduced in the meteorological input file “meteopgt.all”. Instead of wind direction, wind
speed, stability class, and frequency, a different coding is used for the intermediate
files in order to reference them quickly to the original dispersion situation. The coding
is for example:
“65.15,207.1,6,0”
65 is the wind direction in degrees; 15 is the wind speed in cm/s, 207 is the corre-
sponding original dispersion situation, 1 indicates that this is the first intermediate
(after one hour simulation time) flow field of this original dispersion situation, 6 is the
stability class of the original dispersion situation, and the frequency is set to zero (for
numerical reasons in subsequent GRAL dispersion simulations).
This option should be used only for flat or almost flat terrain, whenever
GRAMM wind field simulations are performed to consider the effect of inho-
mogeneous land use (roughness) and vertical wind shifts (Ekman shift) on
subsequent GRAL simulations. Both effects cannot be accounted for when
applying GRAL without GRAMM.
Toolbar:
Back in the GIS form it is possible to check the GRAMM geometry by a 3 D module.
For a better orientation, also at later post processing like the match to observation
function, the receptor points are displayed in the 3D drawing too.
The checkbox “Smooth” creates a commonly used smooth view of the surface. For
quality checks it’s recommended to use the non-smooth view. The vertical scaling
factor should be set to 1 for realistic drawings.
The movement of the camera in the 3D view is done by the cursor-keys (see chapter
19.4).
3D view
The inn-valley nearby Telfs, view angle to northeast, no smooth, the GRAMM grid has a horizontal resolution of 300 m.
To get a feeling about the GRAMM grid frame, you can check the “Show raster”
checkbox in the “Layout manager” (see chap. 11.4) by selecting “GRAMM DOMAIN”
in the “Object manager” (see chap. 11.3).
Further you can define the number of processor cores to be used in the simulations.
The used processor cores per GRAMM computation should not exceed a value of
minimum count of cells per x – or y direction
.
30
If you use multiple instances, the stop and interrupt button and the GRAMM online
feature will work for the first instance only.
Starting GRAMM is done like GRAL (see sec.15.1). After clicking the “Start” button,
the GUI asks for the GRAMM executable. As for GRAL, the GRAMM simulation is
launched with CMD window.
Pressing the “Pause” button interrupts GRAMM and the number of the actual com-
puted dispersion situation is stored in the corresponding field in the “Topography”
tab, while pressing the “Stop” button just interrupts GRAMM.
Start GRAMM
Interrupt GRAMM
Stop GRAMM
If GRAMM wind fields are computed, the GRAMM project is locked to prevent the
project to become invalid. Manual unlocking is possible by clicking in the correspond-
ing lock symbol.
It is recommended to select the desired parameters, and to adjust the graphical de-
sign of the fields to be displayed in the GIS Interface prior to starting GRAMM. Once
GRAMM is running, it is tedious to make changes, because CPUs will be operated at
high load.
For instance, when aiming at visualizing the flow field online, click the box “Wind
vector”. The user will be asked for the vertical layer to be displayed. Then click on the
“Show fields” button to launch the GIS Interface.
All changes made in the GIS Interface during the online session will not be stored.
Before starting GRAMM adjust all visualized fields as desired, because after GRAMM
has been launched it will be tedious to make any change for CPUs will be at full load.
GRAMM updates the fields at every time step. It is possible to add or remove param-
eters for online visualizing during the simulation. To facilitate such actions, it is rec-
ommended to open the CMD window, where GRAMM is running, and to press the
“Pause” key. Then make all changes and afterwards press the “Return” key to con-
tinue the GRAMM simulation.
Besides horizontal fields, vertical profiles can be visualized online, too. Use the cor-
responding button in the GIS Interface, select a field and click on the map, where the
profile should be computed.
Online visualization of
vertical profiles
Toolbar:
The re-order function can be used but is not further developed anymore.
Simulated GRAMM flow fields are based on time series of wind speed, -direction, and stability class for a certain
location in most cases inside the GRAMM model domain. The GUI breaks the time series into some hundred
frequency bins of different dispersion situations. At the end of GRAMM simulations, for each of dispersion situa-
tions, a wind field has been simulated and stored. The Re-Order function searches within these flow fields, which
fits best observed data at the location of the measurement site. It might happen, for instance, that flow field nr.
479 fits best for dispersion situation nr. 1 and so on. In this example, flow field 479 is renamed to 00001.wnd and
stored in the subdirectory “Re-ordered”. This procedure is repeated for all dispersion situations. In this way
significant improvement between simulated and observed wind data at the monitoring site used as input to
GRAMM can be achieved. However, model results should also be checked at other monitoring sites as well, if
available.
The philosophy of the Re-Order function is based on the knowledge that initial- and boundary conditions of a
mesoscale model are never known in all detail, especially in complex terrain. The function searches only for wind
fields with similar stability categories, e.g. a flow field, with a stable dispersion category, is compared only within
other flow fields owing stable classes, too.
The more flow fields have been computed the better the results of the Re-Order function. Therefore, it is best to
compute all flow field situations first, before launching the Re-Order function, although this is not imperative. It
should be noted, that the Re-Order function can also be applied to old GRAMM projects.
Open the GIS-Interface a start the Re-Order function by pressing the button as indicated in the following figure.
Then click on the location of the monitoring site in the map, which has been used as input data for the GRAMM
simulations. The user will be asked for the anemometer height above ground level. The new, re-ordered flow
fields are stored in a new subdirectory “Re-ordered”. The original flow field files are not overwritten but remain in
the subdirectory “Computation”. In order to evaluate the new results after the Re-Order function has been carried
out, it is recommended to create manually another sub-directory (e.g. “Original”) and to copy the original flow field
files (extension *.wnd) to that directory. In a next step, move all flow field files computed via the Re-Order function
from the subdirectory “Re-ordered” to the subdirectory “Computation”. Now the new flow fields can be analyzed
(see sect. 16.10 for more details).
There exist three different types of evaluation procedures (simulations need not to be
finished to use them). In a 1st step open the GIS interface to select an evaluation
method, like in the picture below.
Toolbar:
This function is used to produce a new time series at any user-defined location
or at all receptor points set in the project. Afterwards this time series can be an-
alyzed in the “Meteorology” tab.
Extracting meteorological time series at any location within the model domain
Alternatively, the “Compute wind statistics at all receptor points” checkbox can
be checked. Note that this checkbox is only enabled if receptor points have al-
ready been defined in the project. In this case, wind statistics are calculated au-
tomatically at all receptor points. The heights of the receptor points are used for
the meteorological evaluation.
At the end of the analysis, select the tab “Meteorology” in the main window of
the GUI and use the corresponding buttons to get wind roses and so on. The
Page 132 of 171 User Guide GRAL V20.01
Flow field simulations with GRAMM
file name is indicated in the list box on top, the file itself is stored for re-use in
the subdirectory “Metfiles”. Thus, it can be re-opened at any time the project is
launched again and be analyzed.
If you check this button or the menu entry “Wind field analysis – Analyze stabil-
ity classes” you can select a flow field situation and display the local stability
classes for this flow field situation (available since GRAMM 17.1).
Toolbar:
An average wind speed at a specific height above ground for the time period
can be calculated and visualized by clicking the second button from the left in
the group box or the menu “Wind field analysis – Mean wind speed at a height”.
The result (note that this procedure may also take several minutes) will immedi-
ately be visualized and can be further processed using the object- and layout
managers. The low-pass filter should not be activated in this case.
Toolbar:
Single vector flow fields can be visualized by pressing the third button from the
left in the group box “wind field analysis” or the menu entry “Wind field analysis
– Wind field at a height”.
The user will be asked for the flow field to be shown and the height above
ground. Also, this vector field can be further graphically processed using the ob-
ject- and layout manager (see chapter 11.4).
Vertical profiles of wind speed and –direction for a single flow field
Menu: “Windfield analysis – Vertical profile at a point”
Toolbar:
Select a dispersion situation in the first step. Left mouse clicks at the map loads
the vertical wind profile at this position and the forms with the vertical graphs
are refreshed.
Toolbar:
See chapter 17.1.2.
Toolbar:
Select a dispersion situation and the stability class is displayed as an overlay at
the GIS window.
ggeom.asc (topography)
mettimeseries.dat
meteopgt.all
IIN.dat
optional: landuse.asc
optional: reinit.dat
In case that GRAMM wind fields have already been computed by using the GUI, all
these files can be found in the subdirectory “Computation” of the project.
Toolbar:
Basic concept
In a first step, simulations for a domain can be carried out using not real observation-
al data but artificial wind data comprising a variation of classified situations. Although
simulation times will increase primarily, these wind fields can later be used to match
any new meteorological observations inside the domain.
This function aims at matching already existing GRAMM wind fields to any meteoro-
logical observations inside a GRAMM domain, regardless the period of time when
these measurements have been taken. The imported time series of meteorological
data are synchronized automatically. Thus, it is not necessary to prepare the data in
a way that each time series covers exactly the same time period. Synchronizing is
also possible if one station uses half-hourly mean values and another one is based
on hourly mean values.
The Match-to-Observation function searches the best fitting flow field for each
weather situation of the defined meteorological stations (see chapter 19.5).
The more flow fields are available for the fitting process, the better the results of the
Match-to-Observation function. Therefore, to make the most of this functionality, it is
recommended to compute all flow field situations first, before launching the Match-to-
Observation function, although this is not imperative.
The implementation
After a new project has been defined, import GRAMM wind fields as explained in
sect. 16.11. Then activate the Match-to-Observation procedure (menu or toolbar
button). A new window will open, which allows for importing as many meteo-stations
as desired. Note, that the first (topmost) meteo-station will be used to bin the disper-
sion situations. The order of meteo-stations has no impact on the fitting process
itself, but the stability class is fitted for the first meteo-station only.
Basically, a new meteo-station is imported by clicking the button “Add station”. If this
meteorological file is a *.met file with a coordinate header, the position of that meas-
urement station is set automatically.
If you like to set or change the position of the measurement station, select the meteo
file in the list box and click on the map at the very location of the observation site.
Already imported meteo-stations can be selected in the table by clicking on the sta-
tion.
In doing so, a wind-rose can quickly be depicted (click “Windrose” button), the se-
lected station can be removed from the list (click “Remove” button), or the coordi-
nates of this station can be defined again by clicking on the location on the map. If
you like to add a new station, use the add station button.
The height of the meteo station can be adjusted in the data grid after the meteo file
has been defined.
If all meteo stations are defined, you can start the matching process. The function
reads all computed GRAMM wind fields and stores the wind vectors at all measure-
ment stations.
To improve the quality of matched wind fields, you’ve got several parameters:
Optimization settings
Use local stability Since the GRAMM version 17.1 information about the local
classes stability class for each raster cell will be stored. The prognostic
wind model GRAMM computes the wind vectors and the turbu-
lence in dependence of several parameters, e.g. the topogra-
phy, the land use and the radiation. Therefore, the local wind
vectors at one raster cell will be assigned to a stability class that
differs from the initial wind field. It is recommended to use the
local stability classes to get a reliable coherence between the
local wind vectors and the corresponding stability class.
3x3 average This option uses the average wind vector of a 3x3 raster around
vector the measurement station
SC1/7 weighted This option increases the weighting of the stability classes 1 and
stronger 7
Weighting factor Inside the data grid you can set a weighting factor per observa-
tion station. The higher this value, the stronger the influence of this station to
the optimization algorithm. There is an influence on the weighting of the stability
class error too. The higher the weighting factor, the lower the influence of the
stability class error values (see chapter 19.5).
Direction factor Inside the data grid you can set a weighting factor for the wind
direction per observation station. The higher this value is set, the stronger the
influence of the wind direction of this station. This factor is used at the compo-
nent’s method solely (see chapter 19.5).
Concatenation limit This value allows to reduce the number of flow situations without
losing many details of the flow structures. For all flow situations with a frequen-
cy lower the value “Concatenation limit” a corresponding situation is searched
(see chapter 19.5). A reduced number of flow situations will speed up the GRAL
calculation time and reduce the required disk space.
Note that the concatenation limit does not work entirely in combination with the
sunrise option.
For a detailed description about the matching algorithm look to chapter 19.5
It is possible to save and reload all settings using the respective buttons.
In the next step, the dialog shows how much percent of situations fit within a vectorial
error of 10, 20, 40 and 60 percent and a stability class error of 0 or +-1 classes.
Now it is your decision to finish the matching process or to change the optimization
parameters and repeat the matching procedure. If you repeat the process, the wind
vectors at the meteo positions are already available and the matching process is
much faster than at the first step.
Toolbar:
This function exports a defined part of an existing GRAMM project (including already
computed flow fields) into a new independent GRAMM project.
This new GRAMM project covering only a user-defined part of the original project and
can be used for e.g. subsequent GRAL simulations. It has the advantage of lower
disc-space requirements and faster evaluations and can therefore easily be stored on
computers with low hardware capabilities.
When starting this function, the sub-domain must be defined by clicking and holding
the left mouse button to draw up a rectangle. Thereafter, you need to define the
name of the new GRAMM project and the transfer of data will start automatically.
17. Miscellaneous
17.1. Horizontal and vertical slices for GRAMM and GRAL flow
fields
In the GIS-interface horizontal and vertical slices of GRAMM and GRAL flow fields
can be created by clicking the corresponding buttons, as shown in the following fig-
ure.
Toolbar:
In case of horizontal slices, the user is asked first to select either GRAMM or GRAL.
Note, that only those models are provided for which wind fields have already been
calculated and stored. Horizontal slices are terrain-following at the height specified by
the user.
Horizontal slice
Vertical slice
Toolbar:
In a first step the position of the vertical slice to be performed must be determined by
the user. This is simply accomplished by drawing a line in the map (start with left and
end with right mouse button). Subsequently, a new window opens, where the catego-
rized meteorological situation must be selected in the list in the lower left corner of
the window. Pressing the “Show wind field” button displays the desired flow pattern.
There are "+" and "–" buttons for the vertical zoom, the horizontal zoom and the
length of the wind vectors.
Vertical slices for GRAL surface and buildings need the file “GRAL_geometries.txt“.
GRAL writes that file, if *.gff files have been written (see chapter 8.3). At least one
dispersion situation must be computed to use that option.
GRAL windfields can be displayed, if the corresponding *.gff files for the dispersion
situation exist.
17.1.3 3D view of sections for GRAMM and GRAL surfaces and wind vec-
tors
Use the “3D view” button to show the actual surface and wind vectors as shown in
the “Section drawing” form. The length of the wind vectors is corresponding to the
length in the 2D view of the vertical slice.
The checkbox “Smooth” creates a commonly used smooth view of the surface. For
quality checks it’s recommended to use the non-smooth view.
Toolbar: symbol
Corner points of the polygon to be measured can be set by left clicking on the map.
Measurements are finished with a right-click on the final point or by pressing the ESC
key.
Measure distance
Indicated distance
Toolbox:
Measure area
Indicates area
If the absolute height is lower than the topography surface, the item is automatically
set to the ground level by the GRAL computation core. If you would like to import
items with absolute heights, set the height data at the shapefile to negative absolute
values or use the corresponding checkbox at the import dialog.
Internal, the absolute heights are coded as negative values. Usually the absolute
value is displayed. At the search, select function (see chapter 11.13) the negative
value is shown. That allows the user to filter all items with absolute heights.
Absolute heights will force an error at the GRAL computation core if you try to com-
pute flat terrain without topography (see chapter 17.4).
The difference between absolute and relative heights is shown in the following sec-
tion view at an ascending slope.
To visualize the GRAL topography while editing, it is possible to load the GRAL to-
pography inside the GIS window. Use the menu entry “File – Import – Original GRAL
topography”.
Although you see the GRAL cell heights, the section view (based on the file
“GRAL_topography.txt”) is available, if the GRAL computation has been started and
the 1st situation is computed.
The cell height is visualized in the right corner of the menu bar (see chapter 17.5).
Use the menu entry “Tools – Modify GRAL topography” to fill or strip parts of the
topography. This function is useful to digitize e.g. landfills, gravel quarries or mounds.
The function operates like a “drawing pen” for adding or ablating a relative height or
set an absolute height at the selected cells. The pen-size corresponds to the number
of cells around the mouse cursor affected by this tool.
The absolute height affects cells just once, if the left mouse button is pressed. That
means, if you add +5 m and “paint” several times at the same cell – keeping the
mouse button pressed – only 5 m are added. If you click on a cell multiple times
(release the mouse button while painting) the height changes several times up to the
Hmin/Hmax parameter value.
Using the “absolute height” setting changes the cell height to the specified value
(multiple clicks do not add up).
The absolute height can be limited to a minimum or maximum value by the optional
parameters Hmax and Hmin.
The following example shows the GRAL 3D view of the upper example (left figure)
and a modified topography (right figure) including a soil wall with a relative height of +
10 m and a maximum height of 1180 m.
Smoothing topography is possible using the menu entry “Tools – Low pass filter”
using a 5-point low pass filter. Discard all manual changes using the menu entry
“Tools – Restore GRAL topography”.
Use the menu entry “Tools – Save GRAL topography” to save the modified topogra-
phy. It is not possible to discard the changes, after the modified topography has been
saved.
Sometimes concentration maps, if they are not stored in the default directory
“Maps”
A locked GRAL project will be unlocked immediately, if all *.con or *.grz files are
deleted.
Since the GUI version 19.01, inputs are either “numerical up-downs” or text fields,
using the installed local culture.
For example, in German culture, the “,” is the decimal separator, the “.” a digit group
separator.
In German culture, a value like “25.5” is parsed as 255, but a “25,5” is parsed as
25,5. With German culture settings it is possible to write “1.526,6”, this will be parsed
as 1526,6.
Manual check
If you hit the „store&reload“ button, the inputs are parsed and reloaded. If the process
was successful, you see the input value in your culture (without digit group
separators), otherwise the input will be 0.
Toolbox:
After pressing the button for inserting a north arrow (see next figure), an input field
opens, which asks for the size of the north arrow (default = 100 %). The position can
be adjusted by left clicking on the map. If the position or the size should be changed,
just click again on the button for inserting a north arrow. There is no way to change
the type of arrow.
Adjust size
Insert north arrow
Toolbox:
After pressing the button for inserting a scale bar, an input field opens, asking for the
number of sections of the scale bar. The position of the scale bar can be adjusted by
left clicking on the map. If the position of a scale should be changed, repeat this
procedure. The length, font type, and color can be adjusted by opening the layout
manager (see chapter 11.4). Just double click on the scale bar and the layout man-
ager opens.
Toolbox:
It is possible to save the map as seen on the screen in various pixel graphic formats
(e.g. gif, bmp, etc.). Click on the save button as indicated in the next figure and name
the map to be saved. Afterwards a user will be asked for the desired resolution. Font
sizes may vary according to the chosen resolution.
Saving current
Filename map view
Toolbox:
Using this menu entry or the keyboard sequence Control-C the actual map section is
copied to the clipboard.
19. Appendices
19.1. Methodology „All in all out stables“
Animal farms applying the so-called “all in all out system” do not emit odour at a
constant rate but odour emissions increase from the beginning of each breeding
cycle from zero until they reach a maximum Emax in the end. This is accounted for by
assuming a linear increase in odour emissions for a breeding cycle. Indeed, emis-
sions will rather follow an exponential function than a linear one. Assuming the latter
relationship gives in the following frequency of a particular odour emission during a
breeding cycle:
1
P Normalized frequency of a particular odour emission EGeruch at any day
TZyklus
during the breeding cycle
The probability of an odour hour PGeruch without taking into account other odour
sources in the surroundings, is given by the integral of all frequencies, where the
odour emission results in corresponding odour concentrations in the surroundings
Cgeruch above a pre-defined threshold CGeruchsschwelle :
PGeruch P t C geruch CGeruchsschwelle
TZyklus
If there are other odour sources (e.g. further animal farms) in the surrounding, back-
ground odour concentrations must be considered in the following way:
PGeruch P t C geruch CVorbelastung CGeruchsschwelle
Tgesamt
At last, the frequency of annual odour hours is given at any location in the model
domain by summing up all probabilities of odour perceptions for each hour.
For every receptor point in the model domain the resulting odour concentration from
all of these three different emissions strengths is calculated. There exist seven differ-
ent ways to combine these three emission strengths and frequencies. The possibility
of an odour hour for a certain hour of a year (as defined by the odour threshold con-
centration – e.g. 1 OU/m³) is set to the maximum frequency out of the seven combi-
nations, each owing a different frequency.
Average 2,05
A unit is added following the NODATA value for GRAL concentration, evaluation and
result files. It might be necessary to delete the unit information if the file is used in
other applications.
Accepted row delimiters are the characters blank, tab, and semi-colon. The comma is
allowed as delimiter at the landuse files and for GRAL topography files, if the comma
is not used as decimal separator.
For best compatibility all data values should use the dot as decimal separator. Digit
group separators (for thousands, millions) are not supported.
Accepted row delimiter characters are the blank, comma, tabulator or semi-colon
characters, allowed decimal separators are the dot or the comma. The user can set
the delimiter and decimal character at the import dialog.
The date and time format must use the colon, hyphen or dot character to separate
day, month and year or hour and minute.
For best support of all features, it is recommended to add a 3-line header with the
coordinates of the measurement station (X, Y: coordinate, Z: relative height above
ground).
Example header:
//X=80079
//Y=233949
//Z=10
01.01.2013,00:00, 2.1,218,6
01.01.2013,01:00, 3.2,204,4
…….
19.3.5 DepositionSettings.txt
The rows are tabulator separated. Empty entries are removed. All values must use
the dot as decimal separator.
19.3.6 Emissions_timeseries.txt
Accepted separators are the blank, colon, hyphen, tabulator or semi-colon charac-
ters. Empty entries are removed.
For best compatibility all data values should use the dot as decimal separator. Digit
group separators are not supported.
19.3.7 Precipiation.txt
The accepted separator is the tabulator character. All data values must use the dot
as decimal separator. Digit group separators are not supported.
The number of entries in that file must match the number of entries of the file
mettimeseries.dat.
+ key Zoom in
Right mouse button Set the endpoint of a polyline (e.g. line source), a polygon
(e.g. area source, building) or the portal source
Left mouse button + shift key Set a point and the coordinates manually
Right mouse button + shift key Set an endpoint and the coordinates manually
Left Mouse button + ctrl key Select a point of a line- or area source or a building
Middle Mouse button Keep the button pressed and move the mouse to move the
map
Section drawing
3D module
+ key Zoom in
Cursor left Key and shift key Shift along the X-axis
Cursor right Key and shift key Shift along the X-axis
Cursor left Key and control key Shift along the Z-axis
Cursor right Key and control key Shift along the Z-axis
In a first step, the algorithm reads the computed wind velocity (win), wind directions
(din) and the wind vectors (uin, vin) for each computed weather situation i and the
position of each observation station n.
If the variable SClocal is set, the computed local stability class at the position of the
observation station, otherwise the global stability class of the initial wind field is used
for the value of Sin.
Secondly, the algorithm synchronizes the time series of all observation stations, as
meteorological data doesn’t necessarily need to be provided without data gaps for
the time period in consideration.
Next, the best-fitting computed wind field is picked up from all available wind fields by
the following algorithm:
The measured wind data for each time step M of the time series is read from the
corresponding “*.met” files and the measured values of wind velocity (wMn), wind
direction (dMn), wind vectors (uMn, vMn) for each observation station n and the stability
class of the first observation station are extracted.
For each time step M an error value erri is calculated for each computed weather
situation i and observation station n by the following procedure:
|𝑤𝑖𝑂 − 𝑤𝑀𝑂 |
∗ (𝑓𝑑𝑂 ∗ max{0 | |𝑑𝑖𝑂 − 𝑑𝑀𝑂 | − 12 } 1,8 + ))
max(0.35| min{𝑤𝑖𝑂 | 𝑤𝑀𝑂 })
with 0 ≤ 𝑓𝑑0 ≤ 10
𝑂=1
The lowest value erri indicates the best-fitting computed weather situation for the time
step M.
If the option “remove outliers” is activated (experimental), the inner part of the sum
(here called errin, errmin is the minimum error value for each observation station) is
filtered before the sum
𝑛
∑
𝑂=1
and the error value for each computed situation i is normalized by the number of
summarized observation stations.
As soon as the best-fitting flow field has been found, the frequency of this situation is
incremented by the fraction of time represented by one-time step of the complete
time series. The best fitting situation is written to the file “mettimeseries.dat” using the
date and time stamp from the measurement data and the wind direction, wind speed
and stability class from the initial wind field.
After the matching procedure has been finished for the complete time series, the
computed situations are sorted by their frequency and a new file “meteopgt.all” is
generated.
If the postprocessing - option “Concatenation limit” is set, the algorithm searches for
a well-fitting situation for each wind field with a frequency lower than the value Conc
in permille (at a time series of one year, 1 ‰ represents 9 hours). A situation is con-
sidered “well-fitting”, when it has the same initial wind speed, same initial stability
class, and an initial wind direction within an allowed range of ± 25°. The underlying
idea is that initial wind fields, with almost the same initial conditions, produce quite
similar wind fields within the model domain and, therefore, it is possible to reduce the
number of weather situations without losing too much accuracy. This option leads to
an updating of the “mettimeseries.dat” and “meteopgt.all” files again.
Sometimes the application seems to have no response without a “no response” mes-
sage. In such cases it might happen, that a message box or a notification message is
waiting for a response, but the box is hidden by another (topmost) window. In such
cases try to switch between the windows using the “Alt” + “Tab” keys or try to quit the
message using “Alt” + “O” keys or “Alt” + “F4” keys.
Modulation
"Error when reading modulation factor for ..."
One or more values at the modulation grid are not valid
Source groups
"This number has already been assigned"
You tried to save a source group-number that already exits
Match function
"Error when synchronizing times of meteo stations"
The used meteo station have different timestamps
Georeferencing
"Invalid Input values. Georeferencing failed."
The stretch factor for the bitmap is unrealistic high or low
Dispersion selection
"No situation selected"
Select a situation
Topography tab
"Unable to generate land use file"
Can't write the land use file I/O error
"Topography file is too large. Exceeding available memory space of this computer"
The grids are to large, the needed memory can't be accessed
"Height of the model domain is too low. Increase vertical stretching factor or increase the number of
vertical grid points or increase the height of the first layer"
The top of the model domain needs to be larger than 3 times the maximum elevation
Be careful, overwriting existing wind data may cause in serious problems with your project, if
simulations have already been done
Computation tab
"GRAL simulation stopped or interrupted"
The simulation is finished or has been interrupted by the user or an error
Receptor concentrations
"File zeitreihe.dat not found"
This file might be needed for the computation of time series
"File GRAL_Meteozeitreihe.dat not found"
This file should exist after a finished GRAL-computation with receptors
High Percentiles
"Can´t read meteopgt.all"
This file is needed for GRAL computations and should be in your project computation -folder
"Can´t read mettimeseries.dat"
This file sets the basis data for high percentile computations