FALL3D-7.

USER’S MANUAL

Arnau Folch(1) , Antonio Costa(2) , Giovanni Macedonio(3)

(1)
Barcelona Supercomputing Center (BSC-CNS)
Edifici NEXUS II, 3a planta - c/ Jordi Girona 29
08034 Barcelona, Spain
(2)
Istituto Nazionale di Geofisica e Vulcanologia (INGV)
Via Donato Creti 12 - 40128 Bologna, Italy
(3)
Istituto Nazionale di Geofisica e Vulcanologia (INGV)
Via Diocleziano 328 - 80124 Napoli, Italy

Version release: June 2017

FALL3D-7.2 MANUAL 2

FALL3D-7.2 code Copyright (C) 2017 Arnau Folch, Antonio Costa, Giovanni Macedonio.

This program is free software: you can redistribute it and/or modify it under the terms of the GNU
General Public License as published by the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY;
without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PUR-
POSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not,
visit http://www.gnu.org/licenses/
FALL3D-7.2 MANUAL 3

Contents
1 Introduction 4

2 New features in FALL3D-7.2 4

3 Model equations and parameterizations 5

4 Overview of I/O files and related programs 5

5 Input files and pre-process utility programs 6

5.1 The control file name.inp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.1.1 Block TIME UTC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.1.2 Block GRID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
5.1.3 Block GRANULOMETRY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
5.1.4 Block SOURCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.1.5 Block AGGREGATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
5.1.6 Block AEROSOLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
5.1.7 Block GRAVITY CURRENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5.1.8 Block FALL3D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5.1.9 Block OUTPUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5.1.10 Block POSTPROCESS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
5.2 The TGSD file name.tgsd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
5.2.1 The utility program SETTGSD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
5.3 The meteorological database file name.dbs.nc . . . . . . . . . . . . . . . . . . . . . . . . . 14
5.3.1 The utility program SETDBS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
5.3.2 The utility program GRIB2NC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
5.4 The source file name.src . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.4.1 The utility program SETSRC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.5 The granulometry file name.grn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
5.6 The tracking points file name.pts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

6 Output files and post-process utility programs 19

6.1 The log file name.log . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
6.2 The results file name.res.nc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
6.3 The restart file name.rst.nc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

7 Program installation and execution 20

7.1 Pre-Installation requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
7.2 Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
7.3 Multiple installations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
7.4 Customizing the installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
7.5 Selecting the netCDF library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
7.6 Cleaning directories and uninstalling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
7.7 Model execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Appendices 27
Appendix A. Governing equations and parameterizations . . . . . . . . . . . . . . . . . . . . . 27
Appendix B. Example of control input file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Appendix C. The NetCDF format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Appendix D. Format of the meteo profile file (name.profile) . . . . . . . . . . . . . . . . . . . 39
Appendix E. The GRD format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Appendix F. Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
FALL3D-7.2 MANUAL 4

1 Introduction
FALL3D-7.2 is a 3-D time-dependent Eulerian model for the transport and deposition of tephra. The
model solves a set of advection-diffusion-sedimentation (ADS) equations on a structured terrain-following
grid using a second-order Finite Differences (FD) explicit scheme.
The model inputs are meteorological data, topography, vent coordinate, Eruption Source Parameters
(ESP) such as column height, Mass Flow Rate (MFR), eruption duration, and Total Grain-Size Dis-
tribution (TGSD) which include particle shape and density information. Outputs are tephra ground
load/thickness, airborne ash concentration and other related variables.
The code, written in FORTRAN-90, is available for Unix/Linux/Mac X Operating Systems (OS) and can
be compiled either as serial or parallel using MPI. A set of pre- and post-process utility programs and
related scripts are also included in the FALL3D-7.2 distribution package.
Several parameterizations can be chosen to describe eruption source geometry and physics, particle ter-
minal settling velocity, eddy diffusivity tensor, and ash aggregation.
For the meteorological variables FALL3D-7.2 uses an off-line strategy, i.e. variables are furnished by
independent meteorological models or datasets and interpolated to the FALL3D-7.2 grid as NetCDF files.
The FALL3D-7.2 model can be used to reproduce features of past eruptions, as a tool for short-term ash
dispersal forecasting, and for volcanic fallout hazard assessment.

2 New features in FALL3D-7.2

Relevant changes have been introduced in FALL3D-7. The main new features include:
• A unified source code for both serial and parallel versions (only the serial version was available for
public distribution in previous releases);
• The possibility of using forecasts/reanalysis meteorological data from several global/meso-scale
meteorological models (not available for public distribution in the previous releases);
• Different parameterizations available for ash aggregation. For computational reasons, an initial
Total Grain Size Distribution (TGSD) file (furnished either by the user or generated by the pre-
process SETTGSD utility program) is modified in order to create the final ”effective” granulometry
file filename.grn which can include an aggregated class and a volatile species, treated as tracers;
• An option describing the cloud spreading at the Neutral Buoyancy Level (NBL) for large eruptions
using a semi-analytical gravity current model;

• New parameterizations for resuspension of deposited volcanic ash by wind (based on Folch et al.,
2014);
• A parameterization to account for wet deposition;
• A new empirical parameterization for describing the mass distribution within the volcanic column;

• Two new empirical parameterizations for estimating cross-wind effects on Mass Flow Rate (MFR);
• A new multi-platform installation method that utilizes the configure command option. The pro-
gram can be installed on different machines sharing the same filesystem or on a single machine that
has different compilers.

With respect to FALL3D-7.1 the new version FALL3D-7.2 differs for:

• Added settling velocity model from Dioguardi et al. (2017);
• Upgrade of module settling.f90. Added the WEIBULL and BIWEIBULL distributions in SetTgsd from
Costa et al. (2017);
FALL3D-7.2 MANUAL 5

• GFS scripts modified according to NOAA/NCEP modifications Changes in Grib2nc to allow setting
invert x and invert y from arguments. This affects also SetDbs routines since now all netCDF
files coming from grib decoding are assumed in the lon range (-180,180);
• Script to decode GFS forecasts at 0.25deg added;

• Fixed bug on aggregate settiling velocity correction;

• SetSrc checks if the absolute temperature in the umbrella region remains higher than zero Kelvin
and stops column rise otherwise.
More information and download at:

• http://datasim.ov.ingv.it/fall3d.html or
• http://bsccase02.bsc.es/projects/fall3d/

3 Model equations and parameterizations

The governing equations and the parameterizations used by FALL3D-7.2 are briefly described in the
Appendix A “Governing equations and parameterizations”. For further details see also Costa et al.
(2006); Folch et al. (2009, 2016).

4 Overview of I/O files and related programs

FALL3D-7.2 needs the following input files:
1) The input file name.inp specifying the control parameters and options. This file is read by FALL3D-7.2
and from all utility programs. An example of name.inp is given in Appendix B.

2) The topography and meteorology database file name.dbs.nc. This file is in NetCDF format and is
generated by the pre-process utility program SetDbs (see Section 5.3).
3) The granulometry file name.grn specifying relative fractions and properties of particle classes released
from the source. This file is typically generated by the pre-process utility program SetSrc starting
from the file name.tgsd describing Total Grain Size Distribution (TGSD), density, and shape of
particles. The TGSD file is typically obtained from field data; however it can also be generated by
the utility program SetTgsd (see Section 5.2), assuming either a Gaussian or bi-Gaussian distribution
(in Φ-units). A few options are available for the utility program SetSrc to account for aggregation
effects on fine ash within the eruptive column (see Section 5.4). In this case an extra aggregation class
is added in the name.grn file.

4) The source file name.src specifying the discharge rates at the source points (typically along the
eruptive column). This file is generated by the pre-process utility program SetSrc (see Section 5.4).
5) An optional file specifying a list of points (name.pts) where the tracking of some variables is requested
(e.g. points where to compute tephra arrival times, accumulation rates, etc).
Once a simulation is concluded, FALL3D-7.2 produces the following output files:

1) A log file (name.log) containing information about the run (e.g., summary of input data, error and
warning messages, etc.)
2) The results file (name.res.nc) in NetCDF format (see Appendix C). This file can be processed using
several open-source programs (e.g. ncview, Panoply, ncl, etc.) to generate plots and animations.
Alternatively, the post-process utility program FALL3D2GMT (included in the distribution) can be used
to generate basic GMT scripts automatically.
FALL3D-7.2 MANUAL 6

3) The tracking points files (name.tps.*) containing information about evolution of the variables at the
tracked points. Such information is printed as a single output file for each point specified in the input
file name.pts.
4) A restart file (name.rst.nc) in NetCDF format (see Appendix C). This file is used if the restart
option is on.

A general flowchart of FALL3D-7.2 is shown in Figure 1.

Figure 1: Execution flow of FALL3D-7.2 and related utility programs. The I/O file names are shown in
green/blue.

5 Input files and pre-process utility programs

5.1 The control file name.inp
The control input file in ASCII format consists of a set of blocks defining all the computational and
physical parameters needed by FALL3D-7.2 and its related utility programs (SETTGSD, SETDBS, SETSRC
and FALL3D2GMT). Appendix B shows an example of file name.inp. Each program reads only the necessary
file blocks generating self-consistent input files. Parameters within a block are listed one per record, in
arbitrary order, and optionally can be followed by one (or more) blank space and a comment. The
maximum allowed lenght is 256 characters per line including comments. A detailed description of each
record is given below. Real numbers can be also expressed using the FORTRAN notation (e.g., 12e7 =
12 × 107 ).

5.1.1 Block TIME UTC

This block defines variables related to time, such as the period covered by the meteorology file, beginning
and end of the eruption, etc. It is used by FALL3D-7.2 and by the utility programs SETDBS and SETSRC.
The block has the following format:

• YEAR: Database starting year (YYYY).

• MONTH: Database starting month (MM).
• DAY: Database starting day (DD).
FALL3D-7.2 MANUAL 7

• BEGIN METEO DATA (HOURS AFTER 00): Time (in h after 00UTC of the starting day) at which me-
teorological data start in the database file. This time has to be smaller than time slice defined by
the variables: ERUPTION START (HOURS AFTER 00).
• TIME STEP METEO DATA (MIN): Time step (in min) of the meteorological data in the database file.
The time step can be different from that of the original data (e.g. if the time step is set at 60 min
and the original data were every 6 h, values would be linearly interpolated hourly).
• END METEO DATA (HOURS AFTER 00): Time slice (in h after 00UTC of the starting day) at which
the meteorological data end in the database file. This time has to be larger than time slices defined
by the variables: ERUPTION START (HOURS AFTER 00) and RUN END (HOURS AFTER 00) respectively,
otherwise the program will stop.

• ERUPTION START (HOURS AFTER 00): Time slice of the eruption start (in h after 00UTC of the start-
ing day). These are nt values (nt ≥ 1) indicating the starting times of the different eruptive phases.
Transient behavior of the eruption column can be described by adding a sufficient number of inter-
vals. Eruptive conditions (plume height, MFR, etc.) are assumed constant during each phase. The
first value must be equal or larger than the value of the record BEGIN METEO DATA (HOURS AFTER 00).

• ERUPTION END (HOURS AFTER 00): Time slice of the eruption end (in h after 00UTC of the starting
day). This is the time slice at which the source term is switched off (i.e. the time at which the last
eruptive phase ends).
• RUN END (HOURS AFTER 00): Time slice of the run (in h after 00UTC of the starting day). This
value has to be equal or smaller than the value of the variable
END METEO DATA (HOURS AFTER 00). Note that, in general, a run should continue even after the
source term is switched off (i.e. when the eruption has stopped) in order to allow the remaining
airborne particles to sediment completely.
• RESTART: If YES, the run starts from the restart file name.rst.nc generated at the end of a previous
run.

5.1.2 Block GRID

This block defines the grid variables needed by SETDBS and FALL3D-7.2. The block has the following
format:
• COORDINATES: Map projection options. The two available options are LON-LAT (accounting for
Earth’s curvature) or UTM. It is recommended to use LON-LAT (from version 7.2 the option UTM
is advised not to be used, although still available for backwards compatibility). The UTM option
can only be used if the domain is within a unique UTM zone. The use of the UTM coordinate
system in large domains covering more than one UTM zone is not allowed. The sub-blocks LON LAT
or UTM are read in each case respectively.
• LONMIN: Minimum longitude (in decimal degrees) of the domain (i.e. longitude corresponding to
the bottom left corner). Only used in the LON-LAT option.

• LONMAX: Maximin longitude (in decimal degrees) of the domain (i.e. longitude corresponding to top
right corner). Only used in the LON-LAT option.
• LATMIN: Minimum latitude (in decimal degrees) of the domain (i.e. latitude corresponding to
bottom left corner). Only used in the LON-LAT option.

• LATMAX: Maximin latitude (in decimal degrees) of the domain (i.e. latitude corresponding to top
right corner). Only used in the LON-LAT option.
• LON VENT: Vent longitude. Only used in the LON-LAT option.
• LAT VENT: Vent latitude. Only used in the LON-LAT option.
FALL3D-7.2 MANUAL 8

• UTMZONE: UTM zone code in format nnL (e.g. 33S). Only used in the UTM option.
• XMIN: minimum x-coordinate of the domain (bottom left corner). UTM coordinates must be given
in m. Only used in the UTM option.
• XMAX: maximum x-coordinate of the domain (top right corner). UTM coordinates must be given in
m. Only used in the UTM option.
• YMIN: minimum y-coordinate of the domain (bottom left corner). UTM coordinates must be given
in m. Only used in the UTM option.
• YMAX: maximum y-coordinate of the domain (top right corner). UTM coordinates must be given in
m. Only used in the UTM option.

• X VENT: x-coordinate of the vent. UTM coordinates must be given in m. Only used in the UTM
option.
• Y VENT: y-coordinate of the vent. UTM coordinates must be given in m. Only used in the UTM
option.

• VENT HEIGHT: Height of the vent a.s.l. (in m).

• NX: Number of grid nodes in the x-direction.
• NY: Number of grid nodes in the y-direction.
• ZLAYER (M): Array of heights (in m) of the vertical z-layers in terrain following coordinates. The
vertical layers can be specified manually (as an array of values) or, for equally spaced vertical
discretization, simply indicating the limits and the increment (e.g. FROM 0 TO 10000 INCREMENT
1000). It is not necessary to specify the number of vertical layers since it is automatically calculated.

5.1.3 Block GRANULOMETRY

This block defines the variables needed by the SETTGSD utility program to generate the TGSD file
name.tgsd. The block has the following structure:

• DISTRIBUTION: Type of distribution. The available TGSD are: GAUSSIAN, BIGAUSSIAN, WEIBULL,
BIWEIBULL, where Gaussian, Bi-Gaussian, Weibull or Bi-Weibull distribution refer to distributions
in Φ, where Φ = − log2 d(in mm). For the Weibull distribution see Costa et al. (2016, 2017).
• NUMBER OF CLASSES: Number of granulometric classes in the TGSD. Note that this value can be
different from the number of classes in FALL3D-7.2 (aggregate class and/or volatiles can be added
later in the granulometry file name.grn, created by SETSRC).
• FI RANGE: Minimum and maximum values for the range of Φ considered (Φmin and Φmax respec-
tively).
• DENSITY RANGE: Values of densities ρmin and ρmax (in kg/m3 ) associated to Φ = −1 and Φ = 6
for coarse and fine pumices respectively (e.g Bonadonna and Phillips, 2003). Linear interpolation
is assumed between these two extremes and constant beyond them. In particular, if ρmin = ρmax ,
density will be constant for all classes.
• SPHERICITY RANGE: Values of sphericity ψmin and ψmax associated to Φmin and Φmax particles.
Linear interpolation is assumed between these two extremes and constant beyond them. In partic-
ular, if ψmin = ψmax , sphericity is constant for all classes.

• FI MEAN: Mode of Gaussian distribution in Φ. For Bi-Gaussian distributions two values must be
provided. This value must be provided only for GAUSSIAN or BIGAUSSIAN distributions.
FALL3D-7.2 MANUAL 9

• FI DISP: Standard deviation (σ) of the Gaussian distribution in Φ. For Bi-Gaussian distributions
two values must be provided. This value must be provided only for GAUSSIAN or BIGAUSSIAN
distributions.
• FI SCALE: Shape parameter of Weibull distribution expressed in Φ−unit. For Bi-Weibull distribu-
tions two values must be provided. This value must be provided only for WEIBULL or BIWEIBULL
distributions.
• W SHAPE: Shape factor of the Weibull distribution. For Bi-Weibull distributions two values must be
provided. This value must be provided only for WEIBULL or BIWEIBULL distributions.
• MIXING FACTOR: Only read if distribution type is BIGAUSSIAN or BIWEIBULL. Relative weight p
of grain-size sub-populations, i.e. p for the coarse sub-population and (1 − p) for the fine sub-
population. If this parameter is not specified it is set to the default value of 0.5.

5.1.4 Block SOURCE

This block defines the variables needed by the SETSRC utility program to generate the source term
(eruptive column) for each of the nt ≥ 1 eruptive phases. The block has the following format:
• SOURCE TYPE: Type of source distribution. The available options are POINT, SUZUKI, PLUME or
RESUSPENSION.
1) In the case SOURCE TYPE = POINT only the sub-block POINT SOURCE is used:
• MASS FLOW RATE (KGS): Array of values of the mass flow rate (in kg/s) for the nt eruptive phases.
Alternatively, the user can choose among the options ESTIMATE-MASTIN (Mastin et al., 2009),
ESTIMATE-DEGRUYTER (Degruyter and Bonadonna, 2012) or ESTIMATE-WOODHOUSE (Woodhouse et al.,
2013) and SETSRC automatically computes the MFR from the column heights based on empirical
fits. The last two options account for cross wind effects on plume height and MFR.
• HEIGHT ABOVE VENT (M): Array of column heights (in m above the vent) for the nt eruptive phases.
Note that the plume heights must be lower than the top of the computational domain, specified in
the variable ZLAYER (M) of the GRID block. Otherwise, the program will stop.
2) In the case SOURCE TYPE = SUZUKI only the sub-block SUZUKI SOURCE is used:
• MASS FLOW RATE (KGS): Array of values of the mass flow rate (in kg/s) for the nt eruptive phases.
Alternatively, the user can choose among the options ESTIMATE-MASTIN (Mastin et al., 2009),
ESTIMATE-DEGRUYTER (Degruyter and Bonadonna, 2012) or ESTIMATE-WOODHOUSE (Woodhouse et al.,
2013) and SETSRC automatically computes the MFR from the column heights based on empirical
fits. The last two options account for cross wind effects on plume height and MFR.
• HEIGHT ABOVE VENT (M): Array of column heights (in m above the vent) for the nt eruptive phases.
Note that the plume heights must be lower than the top of the computational domain, specified in
the record ZLAYER (M) of the GRID block. If not, the program will stop.
• A: Array of values of the parameter A in the Suzuki distribution (Pfeiffer et al., 2005) for the nt
eruptive phases.
• L: Array of values of the parameter λ in the Suzuki distribution (Pfeiffer et al., 2005) for the nt
eruptive phases.
3) In the case SOURCE TYPE = HAT only the sub-block HAT SOURCE is used:
• MASS FLOW RATE (KGS): Array of values of the mass flow rate (in kg/s) for the nt eruptive phases.
Alternatively, the user can choose among the options ESTIMATE-MASTIN (Mastin et al., 2009),
ESTIMATE-DEGRUYTER (Degruyter and Bonadonna, 2012) or ESTIMATE-WOODHOUSE (Woodhouse et al.,
2013) and SETSRC automatically computes the MFR from the column heights based on empirical
fits. The last two options account for cross wind effects on plume height and MFR.
FALL3D-7.2 MANUAL 10

• HEIGHT ABOVE VENT (M): Array of column heights (in m above the vent) for the nt eruptive phases.
Note that the plume heights must be lower than the top of the computational domain, specified in
the record ZLAYER (M) of the GRID block. If not, the program will stop.
• THICKNESS (M): Array of values of the thickness of the eruption plume hat for the nt eruptive
phases.

4) In the case SOURCE TYPE = PLUME (based on Folch et al., 2016) only the sub-block PLUME SOURCE
is used:
• SOLVE PLUME FOR: The two available options are MFR or HEIGHT. In the first case SETSRC solves for
the mass flow rate given the column height, whereas in the second case it solves for the column
height agiven the mass flow rate.
• MFR SEARCH RANGE: Two values n and m such that 10n and 10m specify the range of MFR values
admitted in the iterative solving procedure (i.e. it is assumed that 10n ≤ M F R ≤ 10m ). Only
used if SOLVE PLUME FOR=MFR.
• MASS FLOW RATE (KGS): Values of the mass flow rate (in kg/s) for the nt eruptive phases. Only
used if SOLVE PLUME FOR=HEIGHT.
• HEIGHT ABOVE VENT (M): Heights of the plume (in m above the vent) for the nt eruptive phases.
Note that the plume heights must be lower than the top of the computational domain, specified in
the variable ZLAYER (M) of the GRID block. Only used if SOLVE PLUME FOR=MFR.
• EXIT VELOCIY (MS): Values of the magma exit velocity (in m/s) at the vent for the nt eruptive
phases.
• EXIT TEMPERATURE (K): Values of the magma exit temperature (in K) at the vent for the nt eruptive
phases.
• EXIT WATER FRACTION (IN%): Values of the magma volatile fraction (in weight percent) at the vent
for the nt eruptive phases.
• WIND COUPLING: Options are YES or NO. If NO wind effects on the eruptive plume are assumed null;
• AIR MOISTURE: Options are YES or NO. If NO, air specific humidity is assumed zero (dry entrained
air only);

• REENTRAINMENT: Options are YES or NO. If NO particle reentrainment is neglected;

• LATENT HEAT: Options are YES or NO. If NO water latent heat contribution is neglected (note that
the option YES, in some cases, can cause problems of numerical convergence);
• A S: Options for the calculation of radial entrainment coefficient. Possible choices are:

– CONSTANT a s1 a s2. In this case entrainment coefficient is assumed constant using a s1 a s2

for the value of a s in the jet and plume regions respectively. In case a s1 a s2 are not
specified default values of 0.075 and 0.12 are assumed;
– KAMINSKI-R: based on the formulation of Kaminski et al. (2005) as proposed by Folch et al.
(2016) including data of Rouse et al. (1952);
– KAMINSKI-C: based on the formulation of Kaminski et al. (2005) as calculated by Folch et al.
(2016) excluding data of Rouse et al. (1952);
• A V: Options for the calculation of cross-flow entrainment coefficient. Possible choices are:
– CONSTANT a v1. In this case entrainment coefficient is assumed constant using a v1 for the
value of a v. In case a v1 is not specified default values of 0.3 is assumed;
– TATE: based on the formulation of Tate (2002) as proposed by Folch et al. (2016).
FALL3D-7.2 MANUAL 11

4) In the case SOURCE TYPE = RESUSPENSION only the sub-block RESUSPENSION is used:
• MAX RESUSPENSION SIZE (MIC) : Maximum particle size (in µm) for which resuspension is allowed.
This is typically few hundreds of µm.
• DEPOSIT THRESHOLD (KGM2) : Value of the deposit load (in kg/m2 ) encompassing the area where
resuspension is considered. This is used to prevent resuspension in areas with negligible original
deposit loads.
• MAX INJECTION HEIGHT (M): Maximum height of resuspension (in m). Resuspended ash is uniformly
distributed vertically from the ground level to this maximum height.
• EMISSION SCHEME: Type of ash emission scheme (see Appendix A “Governing equations and pa-
rameterizations” for further details). The available options are WESTPHAL (based on Westphal et al.,
1987), MARTICORENA (based on Marticorena and Bergametti, 1995; Marticorena et al., 1997) or SHAO
(based on Shao et al., 1993; Shao and Leslie, 1997; Shao and Lu, 2000).
• EMISSION FACTOR: Tuning factor of the emission scheme.
• THRESHOLD UST: Value of the threshold friction velocity. Only used if EMISSION SCHEME = WESTPHAL
• MOISTURE CORRECTION: If YES, threshold friction velocity is corrected for soil moisture according to
Fecan et al. (1999).

5.1.5 Block AGGREGATION

This block defines the variables needed by SETSRC program in order to modify the original TGSD to
account for ash aggregation effects. It is assumed that one aggregated class is formed. The block has the
following format:
• AGGREGATION MODEL: Aggregation parameterization. Available options are NONE, PERCENTAGE (based
on Sulpizio et al., 2012), CORNELL (based on Cornell et al., 1983) or COSTA (based on Costa et al.,
2010, to be used with the option PLUME only).
• FI AGGREGATES: Φ−class of aggregates (i.e. main mode).
• DENSITY AGGREGATES: Density of aggregates.
• VSET FACTOR: Multiplicative correction factor for settling velocity of aggregates.
• PERCENTAGE (%): Percentage of ash aggregating for classes <FI AGGREGATES.
Only read if AGGREGATION MODEL = PERCENTAGE
• FRACTAL EXPONENT: Fractal exponent (see Costa et al., 2010, for details).
Only read if AGGREGATION MODEL = COSTA

5.1.6 Block AEROSOLS

This block defines the variables needed by SETSRC program in order to add aerosol tracers. This block
defines whether aerosol (indicated as SO2 ) is transported or not and in which percentage (in wt %). In
FALL3D-7.2 only passive transport is considered, i.e. SO2 here is simply considered a tracer and can be
used to simulate the transport of any passive volatile. It is expected to add chemical reactions in future
code releases. The block has the following format:
• SO2: Defines whether SO2 transport is switched on or off. Options are YES (on) or NO (off).
• PERCENTAGE (%): Percentage of SO2 with respect to the total mass.
Note that water is however transported as an additional class when SOURCE TYPE = PLUME consistently
with the water fraction specified in EXIT WATER FRACTION (IN%) in the block SOURCE whereas, for com-
putational simplicity, the mass of the aerosol i.e. SO2 (however typically smaller than a few %) is added
to the total mass, i.e. mass fraction considering SO2 is not normalized.
FALL3D-7.2 MANUAL 12

5.1.7 Block GRAVITY CURRENT

This block defines the variables needed by FALL3D-7.2 in order to describe the effect of volcanic cloud
spreading as a gravity current at the NBL. The model is based on Costa et al. (2013) and parameter
estimations by Suzuki and Koyaguchi (2009). The block has the following format:
• GRAVITY CURRENT: Defines whether the gravity current model is switched on or off.
The available options are YES (on) or NO (off).
• C FLOW RATE: Empirical constant for volumetric flow rate at NBL.
Read only if GRAVITY CURRENT = YES.
• LAMBDA GRAV: Empirical constant for the gravity current model.
Read only if GRAVITY CURRENT = YES.
• K ENTRAIN: Entrainment coefficient for the gravity current model.
Read only if GRAVITY CURRENT = YES.
• BRUNT VAISALA: Frequency of Brunt–Vaisala due to the ambient stratification.
Read only if GRAVITY CURRENT = YES.

5.1.8 Block FALL3D

This block defines the specific variables related to physics in the FALL3D-7.2 program. The block has
the following format:
• TERMINAL VELOCITY MODEL: Type of terminal settling velocity model. The available options are
ARASTOOPOUR, GANSER, WILSON, DELLINO or DIOGUARDI.
• VERTICAL TURBULENCE MODEL: Type of model for vertical diffusion. The available options are
CONSTANT, SIMILARITY or SURFACE LAYER.
• VERTICAL DIFFUSION COEFFICIENT (M2/S): Value of the diffusion coefficient (in m2 /s). Only used
if VERTICAL TURBULENCE MODEL = CONSTANT
• HORIZONTAL TURBULENCE MODEL: Type of model for horizontal diffusion. The available options are
CONSTANT, RAMS, or CMAQ.
• HORIZONTAL DIFFUSION COEFFICIENT (M2/S): Value of the diffusion coefficient (in m2 /s). Only
used if HORIZONTAL TURBULENCE MODEL = CONSTANT.
• RAMS CS: Value of CS in the RAMS model (see eq. 8). Only used if HORIZONTAL TURBULENCE MODEL
= RAMS.
• WET DEPOSITION: Defines whether wet deposition model based on precipation rate is switched on
or off. Options are YES (on) or NO (off).

5.1.9 Block OUTPUT

This block defines specific variables related to output strategy in the FALL3D-7.2 program. The block
has the following structure:
• POSTPROCESS TIME INTERVAL (HOURS): Postprocess time interval in hours.
• POSTPROCESS 3D VARIABLES: The available options are YES or NO. If YES, FALL3D-7.2 writes 3D
concentration in the output file filename.res.nc. If NO, only 2D variables are written to the
output file (this can be desirable for very large files).
• POSTPROCESS CLASSES: The available options are YES or NO. If YES, FALL3D-7.2 writes results for
all classes. If NO, only total results are written. Take into account that the size of the FALL3D-7.2
output file is directly proportional to the number of classes.
• TRACK POINTS: The available options are YES or NO. If YES, FALL3D-7.2 writes the tracking points
files defined in the input file name.pts.
FALL3D-7.2 MANUAL 13

5.1.10 Block POSTPROCESS

This block of data is read by the post-process utility program FALL3D2GMT, which writes a script used to
post-process results using GMT. The block has the following format:
• CROP DOMAIN: Values for LONMIN, LONMAX, LATMIN, and LATMIN of the GMT domain. Note than the
domain in GMT (domain of the plot) can be smaller than the simulation domain.
• MAP TOPOGRAPHY: The available options are YES or NO. If YES, FALL3D2GMT plots topography contours
(in m) as specified in the corresponding CONTOUR LEVELS line.
• MAP TOTAL LOAD: The available options are YES or NO. If YES, FALL3D2GMT plots contours of total
deposition load (in kg/m2 ) as specified in the corresponding CONTOUR LEVELS line.
• MAP WET LOAD: The available options are YES or NO. If YES, FALL3D2GMT plots contours of total wet
deposition load (in kg/m2 ) as specified in the corresponding CONTOUR LEVELS line. Note that this
is possible only when WET DEPOSITION is set to YES.
• MAP CLASS LOAD: The available options are YES or NO. If YES, FALL3D2GMT plots contours of class
deposition load (in kg/m2 ) as specified in the corresponding CONTOUR LEVELS line. Note that this
is possible only when POSTPROCESS CLASSES is set to YES.
• MAP CLASS WET: The available options are YES or NO. If YES, FALL3D2GMT plots contours of class wet
deposition load (in kg/m2 ) as specified in the corresponding CONTOUR LEVELS line. Note that this
is possible only when WET DEPOSITION and POSTPROCESS CLASSES are set to YES.
• MAP DEPOSIT THICKNESS: The available options are YES or NO. If YES, FALL3D2GMT plots contours of
total deposit thickness converted using the density value specified in the DEPOSIT DENSITY (KG/M3)
line (by default set to 1000). Values can be expressed in mm, cm, or m in accord to the UNITS line
for the contours specified in the corresponding CONTOUR LEVELS line.
• MAP CONCE GROUND: The available options are YES or NO. If YES, FALL3D2GMT plots contours of total
concentration at ground level (in g/m3 ) as specified in the corresponding CONTOUR LEVELS line.
• MAP PMxx GROUND: The available options are YES or NO. If YES, FALL3D2GMT plots contours of P M5 ,
P M10 , and P M20 concentrations at ground level (in g/m3 ) as specified in the corresponding
CONTOUR LEVELS line.
• MAP COLUMN MASS: The available options are YES or NO. If YES, FALL3D2GMT plots contours of total
column mass (in g/m2 ) as specified in the corresponding CONTOUR LEVELS line.
• MAP COLUMN PMxx: The available options are YES or NO. If YES, FALL3D2GMT plots contours of P M5 ,
P M10 , and P M20 column mass (in g/m2 ) as specified in the corresponding CONTOUR LEVELS line.
These are useful quantities to compare with satellite imagery.
• MAP FLIGHT LEVEL:The available options are YES or NO. If YES, FALL3D2GMT plots contours of con-
centration at different Flight Levels (in g/m3 ) as specified in the corresponding CONTOUR LEVELS
line. By default, these are FL050, FL100, FL150, FL200, FL250, FL300, FL350 and FL400. How-
ever, the number and values of the different FL can be easily modified (just edit the FALL3D-7.2
source file InpOut.f90, modify nflevel as desired and re-complie the code).
• MAP AOD: The available options are YES or NO. If YES, FALL3D2GMT plots contours Aerosol Optical
Depth as specified in the corresponding CONTOUR LEVELS line.

5.2 The TGSD file name.tgsd

The TGSD file is an ASCII file containing the definition of the particle classes (a class is characterized
by particle size, density and sphericity) without aggregates nor aerosol components. Note that in the
previous versions of the code this file was called name.grn. The file format is described in Table 1 and
the meaning of the used symbols is the following:
FALL3D-7.2 MANUAL 14

• nc: Number of particle classes.

• diam: Class diameter (in mm).
• rho: Class density (in kg/m3 ).

• sphe: Class sphericity.

P
• fc: Class mass fraction (0-1). It must verify that fc = 1.

nc
diam(1) rho(1) sphe(1) fc(1)
...
diam(nc) rho(nc) sphe(1) fc(nc)

Table 1: Format of the TGSD file name.tgsd.

5.2.1 The utility program SETTGSD

The utility program SETTGSD can be used to generate the TGSD file name.tgsd in accordance with the
parameters specified in the block GRANULOMETRY of the input file name.inp. This program generates only
Gaussian and Bi-Gaussian distributions in Φ (log-normal in d) and assumes a linear increase of density ρ
and sphericity ψ between the two extremes specified by the user. For other grain size distributions, the
user must provide the TGSD file (typically derived from field data). Note that in the previous versions
of the code this utility program was called SETGRN.

5.3 The meteorological database file name.dbs.nc

The file name.dbs.nc, written in NetCDF format (see Appendix C), contains topography and time-
dependent meteorological data needed by FALL3D-7.2 (wind field, air temperature and density, humidity,
etc), written in terrain-following coordinates. The file is created by the utility program SETDBS, which
interpolates meteorological data from the original grid of meteorological models to the FALL3D-7.2 com-
putational domain. In turn, SETDBS requires a specific naming convention for the original NetCDF
meteorological data. For this reason, another utility program (GRIB2NC) can be used in combination with
wgrib/wgrib2 to decode and convert original grib-format meteorological data to the required NetCDF
format. There are a several options to generate this database file depending on the scale of application.

5.3.1 The utility program SETDBS

The pre-process utility program SETDBS generates the database file name.dbs.nc in accordance with the
parameters specified in the blocks TIME UTC and GRID of the input file name.inp. After running the utility
GRIB2NC , the program SETDBS can use meteorological data from different independent meteorological
models and interpolates variables onto the FALL3D-7.2 computational grid. The time duration of the
database must be equal or larger than the duration of the simulation. The possible options are listed in
Table 2 and described below.

• The simplest option consists of using a horizontally uniform wind derived from a single vertical
profile, typically obtained from sounding measurements or from indirect reconstructions. The ver-
tical profile needs to be specified in the ASCII file name.profile using the format described in the
Appendix D. In this case, in addition to the profile name.profile it is also necessary to furnish
a topography file name.top in GRD format (see Appendix E). It is recommended to use this very
simplistic option (homogeneous wind field) only when no other meteorological data are available.
FALL3D-7.2 MANUAL 15

Model Native data format need of GRIB2NC

Global model forecasts
GFS at 1o resolution grib yes
GFS at 0.5o resolution grib yes
Global analysis and re-analysis
NCEP final analysis at 1o resolution grib yes
NCEP re-analysis 1 grib no
NCEP re-analysis 2 grib no
ECMWF ERA-40 re-analysis grib yes
ECMWF ERA-Interim re-analysis grib yes
Mesoscale models
WRF-ARW NetCDF no
ETA grib yes
ARPA-SIM grib yes
Others (local-scale options)
CALMET-6.2 own-binary no
Vertical profile ASCII no

Table 2: List of different options handled by the SETDBS pre-process utility program. Native data in grib
format needs to be converted first to NetCDF using the GRIB2NC program before running SETDBS.

• The second choice (CALMET option) uses data derived from the output of the meteorological diagnos-
tic model CALMET (Scire et al., 2000). This option is typically used for assimilating and interpolating
short-term forecasts (or re-analysis) from Mesoscale Meteorological Prognostic Models (MMPM)
to a finer scale. In this case only the UTM coordinate system can be used. Note that the output
of CALMET is a binary file that depends on the architecture of the machine where it was generated.
Moreover note that this option is compatible only with a CALMET output time step equal to an hour
(i.e., nsecdt=3600).
• The third choice (strongly recommended) uses data from global/mesoscale forecasts or analysis/re-
analysis. Global data (assumed to be downloaded in grib format) have to be decoded and converted
to NetCDF using the utility program GRIB2NC before running SETDBS.

5.3.2 The utility program GRIB2NC

The utility program GRIB2NC runs in combination with wgrib/wgrib2 and dedicated scripts (see Section
7.7) in order to convert grib format data to a NetCDF file readable by SETDBS. Possible options are:

1. GFS forecasts at 1o and 0.5o resolution.

• http://nomad3.ncep.noaa.gov/ncep data/index.html
• http://www.nco.ncep.noaa.gov/pmb/products/gfs
• 26 pressure levels starting at 10 mb ( 10 20 30 50 70 100 150 200 250 300 350 400 450 500 550
600 650 700 750 800 850 900 925 950 975 1000).
• 7 days ahead (00 cycle assumed).
2. NCEP GFS final analysis (FNL) at 1o resolution. This product is from the Global Data Assimilation
System (GDAS). The FNLs are made with the same model which NCEP uses in the Global Forecast
System (GFS), but the FNLs are prepared about an hour or so after the GFS is initialized.
• http://rda.ucar.edu/datasets/ds083.2/
• 26 pressure levels starting at 10 mb (10 20 30 50 70 100 150 200 250 300 350 400 450 500 550
600 650 700 750 800 850 900 925 950 975 1000).
FALL3D-7.2 MANUAL 16

• 4-times daily.
• Temporal range: 30 July 1999 to present.
3. CDAS-NCEP/NCAR reanalysis 1 at 2.5o resolution.
• http://nomad3.ncep.noaa.gov/ncep data/index.html
• 17 pressure levels starting at 10 mb (10 20 30 50 70 100 150 200 250 300 400 500 600 700 850
925 1000).
• 4-times daily.
• Temporal range: 1 January 1948 to present.
• NOTE: only pressure levels 00 cycle grib files (pgb.ft00.YYYYMM) are used.
4. NCEP Reanalysis 2 at 2.5o resolution.
• http://nomad3.ncep.noaa.gov/ncep data/index.html
• 17 pressure levels starting at 10 mb (10 20 30 50 70 100 150 200 250 300 400 500 600 700 850
925 1000).
• 4-times daily.
• Temporal range: 1 January 1979 to present.
• NOTE: only pressure levels grib files (pgb.YYYYMM) are used.
5. ECMWF ERA-40 reanalysis.

• http://apps.ecmwf.int/datasets/data/era40 daily
• 23 pressure levels starting at 1 mb (1 2 3 5 7 10 20 30 50 70 100 150 200 250 300 400 500 600
700 775 850 925 1000).
• 4-times daily.
• Temporal range: 1 September 1957 to 31 August 2002.
• NOTE 1: Files are downloaded separately for pressure, non-pressure (surface) and invariant
variables. Required variables are: i) pressure: Geopotential, Relative humidity, Tempera-
ture, U component of wind, V component of wind, Vertical velocity, ii) Surface: 10 me-
tre U wind component, 10 metre V wind component, 2 metre temperature, Boundary layer
height, iii) Invariant: Geopotential, Land-sea mask. Variable files must be named with stan-
dard ECMWF names, i.e. hgt.grib, temp.grib, rhum.grib, omega.grib, uvel.grib,
vvel.grib, t2sfc.grib, u10sfc.grib, v10sfc.grib,.
• NOTE 2: Downloading a domain subset with different resolutions (lower than 0.25o ) is possible.
6. ECMWF ERA-Interim reanalysis.

• http://apps.ecmwf.int/datasets/data/interim full daily

• 37 pressure levels starting at 1 mb (1 2 3 5 7 10 20 30 50 70 100 125 150 175 200 225 250 300
350 400 450 500 550 600 650 700 750 775 800 825 850 875 900 925 950 975 1000).
• 4-times daily.
• Temporal range: 1 January 1979 to present.
• NOTE 1: Files are downoladed separately for pressure, non-pressure (surface) and invariant
variables. Required variables are: i) pressure: Geopotential, Relative humidity, Temperature,
U component of wind, V component of wind, Vertical velocity, ii) Surface: 10 metre U wind
component, 10 metre V wind component, 2 metre temperature, Boundary layer height, iii)
Invariant: Geopotential, Land-sea mask.
• NOTE 2: Downloading a domain subset with different resolutions (lower than 0.25o ) is possible.
FALL3D-7.2 MANUAL 17

5.4 The source file name.src

The source file name.src is an ASCII file containing the definition of the source term. The source can
be defined for different time phases during which source values are kept constant. The number, position
and values (i.e. Mass Flow Rate) of the source points can vary from one time slice to another and cannot
overlap. There is no restriction on the number and duration of the time slices. It allows, in practice,
to discretize any kind of source term. This file can be defined directly by the user or generated by the
pre-process utility program SETSRC. The format of the file name.src is described in Table 3 and the
meaning of the used symbols is the following:
• itime1: Source starting time (in sec after 00UTC of the eruption starting day).

• itime2: End time (in sec after 00UTC of the eruption starting day).
• nsrc: Number of source points (it can vary from one interval to another depending on the column
height).
• nc: Total number of particle classes (as in the file name.grn).

• MFR: Mass flow rate (in kg/s).

• x: Longitude or x-coordinate of the source isrc.
• y: Latitude or y-coordinate of the source isrc.
• z: z-coordinate of the source point above ground level (a.g.l.) (in m).

• i,j,k: indexes of the source point in the FALL3D-7.2 mesh.

• src: P
Mass
P flow rate (in kg/s) of each granulometric class for this point source. It must be verified
that src(isrc, ic) = M F R.

itime1 itime2
nsrc nc
MFR
x y z i j k src(1,1) ... src(1,nc)
...
x y z i j k src(nsrc,1) ... src(nsrc,nc)

Table 3: Format of the source file name.src. This block is repeated for each eruption phase.

5.4.1 The utility program SETSRC

The utility program SETSRC is used to generate 1) the source file name.src in accordance with the
parameters specified in the blocks TIME UTC and SOURCE of the input file name.inp and 2) modify the
preliminary TGSD file name.tgsd in order to create the granulometry file name.grn used by FALL3D-7.2
and consistent with the aggregation and aerosol options specified in the blocks AGGREGATION and AEROSOLS
of the control file name.inp.
Available options are i) a point source column, ii) a mushroom-like shape column (Suzuki option), iii)
an eruption column model based on the Buoyant Plume Theory (BPT) or iv) a diffuse emission for ash
resuspension.
FALL3D-7.2 MANUAL 18

5.5 The granulometry file name.grn

The granulometry file is an ASCII file containing the characterization of the particle classes and, option-
ally, aggregates and aerosols. This file is created by the utility program SETSRC from the preliminary
TGSD file name.tgsd. Note that this file is different from the one used in the previous versions of the
code. The file format is described in Table 4 and the meaning of the used symbols is the following:
• nc: Total number of particle classes (differs from nc in the TGSD file in case of aggregation or
aerosols).
• diam: Class diameter (in mm).

• rho: Class density (in kg/m3 ).

• sphe: Class sphericity.
P
• fc: Class mass fraction (0-1). It must verify that fc = 1.
• class: Label denoting class (particle/gas typology) such particle class, aggregate or gas.

nc
diam(1) rho(1) sphe(1) fc(1) class(1) (e.g. class-01)
... ... ... ... ...
diam(nc-2) rho(nc-2) sphe(nc-2) fc(nc-2) class(nc-2) (e.g. aggregate)
diam(nc-1) rho(nc-1) sphe(nc-1) fc(nc-1) class(nc-2) (e.g. H2O)
diam(nc) rho(nc) sphe(nc) fc(nc) class(nc-2) (e.g. SO2)

Table 4: Format of the granulometry file name.grn

Note that when the option COSTA is selected as the aggregation model, the variable fc is automatically
calculated by setsrc since it can vary with time.

5.6 The tracking points file name.pts

This file contains the names (identifiers) and coordinates of the points to be tracked. It is used only
when the record TRACK POINTS in the input file name.inp is set to YES. The format of the file name.pts
consists of lines (one line per point) with three columns specifying the point name, the point longitude (or
x-coordinate if UTM coordinates are used), and the point latitude (or y-coordinate if UTM coordinates
are used). If available, an extra column with measurements of tephra loading on the ground can be added
for a sake of comparison with simulation results. There is no limit on the number of points to track. The
file format is described in Table 5.

location-1 x-coord-1 x-coord-1 measurement-1

location-2 x-coord-2 x-coord-2 measurement-2
location-3 x-coord-3 x-coord-3 measurement-3
location-4 x-coord-4 x-coord-4 measurement-4
... ... ... ...
location-n x-coord-n x-coord-n measurement-n

Table 5: Format of the granulometry file name.pts

FALL3D-7.2 MANUAL 19

6 Output files and post-process utility programs

6.1 The log file name.log
The file name.log is an ASCII file where critical information about the simulation run is stored. The
information written on the name.log contains:
• FALL3D-7.2 copyright;
• code version;
• number of processors;
• starting time of the simulation;
• input files names and paths;
• output files names and paths;
• time range and grid data of the meteorological database;
• FALL3D-7.2 input data (e.g. time range; numerical parameters; output options; etc);
• memory requirements;
• source terms features;
• particle classes;
• atmospheric properties, horizontal, and vertical diffusion;
• terminal velocities;
• main parameters of the gravity current model;
• updates about the simulation such as iteration number, critical time step, elapsed time, current
(simulation time), and a mass balance for the total mass (inside and ouside the computational
domain) and the erupted mass.

6.2 The results file name.res.nc

This file, written in netCDF format, contains the following output variables:
• particle properties (diameter, density and sphericity);
• topography;
• ground load and, if specified in the control input file, class ground load;
• wet deposition and, if specified in the control input file, class wet deposition;
• deposit thickness;
• total and P Mxx (xx=5, 10, 20) concentration at ground level;
• total and P Mxx (xx=5, 10, 20) column mas load (vertical integration of concentration);
• concentration at different flight levels. By default, these are FL050, FL100, FL150, FL200, FL250,
FL300, FL350, and FL400. However, these values can be reconfigured by modifying the InpOut.f90
source file and recompiling the code;
• aerosol optical depth;
• total and class concentration at all model layers (if specified in the control input file only).
FALL3D-7.2 MANUAL 20

6.3 The restart file name.rst.nc

The restart file, written in netCDF format, can be used to start a new run from the end of a previous
simulation. The file is automatically created each time FALL3D-7.2 prints its results. If RESTART = YES
in the block TIME UTC of name.inp, a run is initialized with the airborne concentration specified in the
restart file name.rst.nc. Any restart file previously created is destroyed whenever a new restart file is
printed.

7 Program installation and execution

IMPORTANT NOTE: Please read very carefully all this section before installing and running FALL3D-7.2.

7.1 Pre-Installation requirements

FALL3D-7.2 is written in FORTRAN-90 and requires the external library netCDF. The code has been tested
in UNIX/Linux platforms including MacOS X. The source code can be compiled to run in serial mode
(one processor) and parallel mode (multiple processor and/or clusters based on MPI). For compiling the
serial version you will need:
• A Fortran 90 compiler (eg: ifort, gfortran, xlf90, etc.)
• The Library netCDF installed on your machine (versions 3.6.*, 4.* or later). NetCDF library is
available from: http://www.unidata.ucar.edu/netcdf/
In addition, for compiling the parallel version (optional), you will need:
• MPI or OpenMPI installed on you machine with Fortran compilation enabled (command mpif90
must be available on your machine).

To decode meteorological data in grib format (see Table2) using GRIB2NC you will need:
• wgrib and/or wgrib2, available from http://www.cpc.ncep.noaa.gov/products/wesley/wgrib.html
Finally, if you wish to use post-process results using the utility program FALL3D2GMT (optional) you will
need:

• The Generic Mapping Tools (GMT) library, available from

http://gmt.soest.hawaii.edu/projects/gmt/wiki/Download

7.2 Installation
Sources of FALL3D-7.2 are distributed as a gzipped tar archive in a file named fall3d-7.2.tar.gz,
available at the following URLs:
http://bsccase02.bsc.es/projects/fall3d/
http://datasim.ov.ingv.it/fall3d.html
After you obtained file fall3d-7.2.tar.gz, copy it in a directory (eg: your home directory) and unpack
the tarball with the command:
$ tar zxvf fall3d-7.2.tar.gz
This creates the installation directory tree with the root directory named fall3d-7.2. The directory
tree is shown in Table 6.
The package comes with a configure script for automatically configuring your installation.
The ‘configure’ shell script attempts to guess correct values for various system-dependent variables
used during compilation. Then uses these values to create a ‘Makefile’ in each directory of the package
for compiling and installing the code and the scripts.
FALL3D-7.2 MANUAL 21

Table 6: Default structure of fall3d-7.2 sub-folders.

Manual # Contains this manual
Example # Contains an example of input file (name.inp)
libMaster # Master library for Fall3d and utilities
Utilities # Utilities root directory
Utilities/Grib2nc # Grib2nc sources
Utilities/Grib2nc/config # Grib2nc sample configuration files
Utilities/SetTgsd # SetTgsd sources
Utilities/SetDbs # SetDbs sources
Utilities/ConfigScripts # Shell script sources
Utilities/Fall3d2GMT # Fall3d2GMT sources
Utilities/SetSrc # SetSrc sources
Scripts # This is generated by ’make install’
bin # This is generated by ’make install’

The configuration script does not automatically set the compiler flags. It is strongly suggested to set
the compiler optimization flag “-O” by setting the environmental variable FCFLAGS=-O or by setting
it in the configure command line, as shown below.
In brief, to configure, compile and install FALL3D-7.2 (serial version) it should be enough to issue the
following commands:
$ cd fall3d-7.2
$ ./configure FCFLAGS=-O
$ make
$ make install
For installing both the serial and the parallel versions the commands are:
$ cd fall3d-7.2
$ ./configure FCFLAGS=-O --enable-parallel
$ make
$ make install
By default, the binary files are installed in the directory ‘bin’ and the shell scripts (used to launch
FALL3D-7.2 and all the related utility programs) in the directory ‘Scripts’, both under the root directory
fall3d-7.2. Please note that the shell scripts located in directory ‘Scripts’ are generated by ‘make
install’ and contain pointers to the location of the binaries. The shell scripts can be moved/copied to
other directories, but the binary files must be left in the ‘bin’ directory. The default location of the bin
directory can be changed by providing proper flags to the ‘configure’ script, as described below (see
flag --prefix or --exec-prefix). The variable SCRIPTDIR, can be changed by passing its value to the
command line of configure (configure SCRIPTDIR=directory).
In the default configuration, the meteo data are searched in the directory ‘$HOME/Data’ and the
FALL3D-7.2 runs are stored in the directory ‘$HOME/Runs’. However, these directories can be modified
as described below.

7.3 Multiple installations

Multiple installations of FALL3D-7.2 on the same system might be needed. This is usually the case
when a disk is shared among different computer platforms and/or when you want different versions of
FALL3D-7.2 compiled for different platforms or on a single platform but with different libraries and
compilers. To do this, just untar file ‘fall3d-7.2.tar.gz’ into different directories and configure each
replica of FALL3D-7.2 using different configuration flags. In this case, the binaries may reside in different
directories, but can share the same data file (eg: the same ‘DATADIR’ and ‘RUNDIR’ directories). The
different binaries can be launched by the corresponding shell scripts (located in the directory ‘SCRIPTDIR’).
FALL3D-7.2 MANUAL 22

7.4 Customizing the installation

You may choose different configuration parameters, including the fortran compiler, compilation flags, the
netCDF version (if you have more than one) and the location of the directories ‘bin’, Scripts’, ‘Data’
and ‘Runs’. The values of the different variables and flags are printed on the screen by ‘configure’ at
the end of the procedure (if no errors occur). During the configuration, you should pay attention
to the information printed by the command ‘configure’.
A typical output of the configure script is shown in Table 7, where the list of the variables that
are set during the installation is shown. The user settable variables are: FC, MPIF90, MPIEXEC, FCFLAGS,
F77, NETCDF, NC INC, NC LIB, WGRIB, WGRIB2, RUNDIR, DATADIR and SCRIPTDIR. These variables can be
changed by defining them as shell environment variables or by passing their value as argument in the
’configure’ command line. In case you specify a variable both in the environment and in the command
line, the value passed in the command line takes the precedence.
Example (see Table 7): the default Fortran compiler found by configure is FC=gfortran. You can
change the default value in the following ways:
(Bourne shell)
$ export FC=ifort
$ export FCFLAGS=-O
$ ./configure
or
(C-shell)
$ setenv FC ifort
$ setenv FCFLAGS -O
$ ./configure
or
(any shell)
$ ./configure FC=ifort FCFLAGS=-O
All the variables listed before can be configured in a similar way. For example, you can set the run
directory (RUNDIR) and the data directory (DATADIR) to different values:
$ ./configure DATADIR=/home/myself/fall3d/mydata RUNDIR=/home/myself/fall3d/myruns
Moreover you can specify the installation prefix (root directory for the installation of the binaries) with
the flag --prefix=DIRECTORY and/or choose to install also the parallel version. Example:
$ ./configure --prefix=/home/myself/fall3d --enable-parallel FCFLAGS=-O [...]

7.5 Selecting the netCDF library

The path of the netCDF include files and libraries is defined by the variable NETCDF, which specifies the
root directory of the netCDF installation. In particular, the include files (eg: file netcdf.inc) are stored
in the directory $NETCDF/include and the libraries (eg: file libnetcdf.a) are stored in $NETCDF/lib.
The configure script should automatically set the proper value of the NETCDF path. This path is searched
using a guess algorithm. First of all ’configure’ looks for the program ’nc-config’, provided with the
latest versions of netCDF; if it is found, nc-config is called and the variables NETCDF, NC INC and NC LIB
are set accordingly. Otherwise, the NETCDF path is defined by the location of the program ncdump, usually
stored in the directory $NETCDF/bin. In case ’configure’ is not able to locate your netCDF installation,
or you want to select another version of netCDF, you can manually set the variable NETCDF. This cand be
done, as for the other variables, either by passing them as an environmental variable, or in the ’configure’
command line:
$ ./configure NETCDF=/usr/local/netcdf
FALL3D-7.2 MANUAL 23

Table 7: Typical output of the configure script

[... configure messages ...]

configure: Configuration complete - Fall3d-7.2 (serial)
configure:
configure: Using netCDF 4.2.1
configure: Fortran 90 compiler: FC=gfortran
configure: Enable parallel version: --enable-parallel=no
configure: Fortran 90 PAR compiler: MPIF90=(not used)
configure: Launcher of MPI programs: MPIEXEC=(not used)
configure: Fortran flags: FCFLAGS=-O
configure: Fortran 77 compiler: F77=gfortran
configure: Root directory of netcdf: NETCDF=/usr/local
configure: Compiler flags for netcdf: NC_INC=-g -I/usr/local/include
configure: Linker flags for netcdf: NC_LIB=-L/usr/local/lib -lnetcdff -lnetcdf
configure: Grib files reader WGRIB=wgrib
configure: Grib2 files reader WGRIB2=wgrib2
configure: Grib2nc config directory: GRIBCONFDIR=[...]/Utilities/Grib2nc/config
configure: Install prefix: --prefix=[...]/fall3d-7.2
configure: Executables install prefix: --exec_prefix=${prefix}
configure: Binary directory: --bindir=${exec_prefix}/bin
configure: Run directory: RUNDIR=$HOME/Runs
configure: Data directory: DATADIR=$HOME/Data
configure: Scripts directory: SCRIPTDIR=[...]/Scripts

The variables NC INC and NC LIB represent the flags passed, respectively, to the fortran compiler and to
the linker, in order to use netCDF (see Table 7 for the typical values). Usually, NC INC and NC LIB are
inferred from the value of NETCDF.
The variable GRIBCONFDIR represents the directory where the configuration files of the utility Grib2nc
are stored. Just as the other variables, it can be changed by passing a new value to the ./configure
command line or by setting the corresponding environment variable.

7.6 Cleaning directories and uninstalling

After compilation and installation, you can remove the program binaries and object files from the source
directory by typing ‘make clean’. To remove the files that ‘configure’ has created (so you can compile
the package for a different configuration), type ‘make distclean’. This command does not remove the
binaries and the scripts installed in the directory SCRIPTDIR.
To uninstall FALL3D-7.2 you can type ‘make uninstall’ to remove the binary files in the directory
$exec prefix/bin and the scripts in the directory SCRIPTDIR. It does not remove the source files.

7.7 Model execution

To run a new simulation named name simply create a new directory (or a symbolic link) called name in
the folder Runs and create a new control input file name.inp (or simply copy the file Example.inp located
in the folder Example, rename it as name.inp and modify it). FALL3D-7.2 and the utility programs are
launched using a serie of scripts with some arguments (an option is to create alias for the scripts, so that
these can be called directly from any folder). The execution flow is as follows:
1. First, place the meteorological data files in the appropriate folder (or create a symbolic link to data
in this folder):

• Data/model-grib for meteo files in grib format or,

FALL3D-7.2 MANUAL 24

• Data/model-nc for meteo files in NetCDF format

where model take one of the following names (see Table 2):
• gfs05deg for GFS forecasts at 0.5o resolution.
• gfs1deg for GFS forecasts at 1o resolution.
• ncepFNL for NCEP GFS final analysis (FNL) at 1o resolution.
• ncep1 for CDAS-NCEP/NCAR reanalysis 1 at 2.5o resolution.
• ncep2 for NCEP Reanalysis 2 at 2.5o resolution.
• era40 for ECMWF ERA-40 reanalysis.
• eraIn for ECMWF ERA-Interim reanalysis.
• wrf for WRF-ARW output.
• eta for ETA output.
• arpa for ARPA-SIM output.
• calmet62 for CALMET (version 6.2) output.
• profile for vertical profile.
NOTE: The scripts assume that files in folder Data/model-grib are named name.model.grib and
files in folder Data/model-nc are named name.model.nc (e.g. name.wrf.nc for WRF-ARW output
files located in folder Data/wrf-nc, name.eraIn.grib for ERA-Interim grib files located in folder
Data/eraIn-grib, etc).
2. If necessary (only for meteo files in grib format), run wgrib/wgrib2 and GRIB2NC. For this, you
have to launch one of the following scripts:
• Script-gfs05deg-to-nc name
where name is the name of the current run. This decodes the file name.gfs05deg.grib (GFS
forecasts at 0.5o resolution), placed in the folder Data/gfs05deg-grib, and creates the file
name.gfs05deg.nc in the folder Data/gfs05deg-nc for subsequent execution of SETDBS.
• Script-gfs1deg-to-nc name
where name is the name of the current run. This decodes the file name.gfs1deg.grib (GFS
forecasts at 1o resolution), placed in the folder Data/gfs1deg-grib, and creates the file
name.gfs1deg.nc in the folder Data/gfs1deg-nc for subsequent execution of SETDBS.
• Script-ncepFNL-to-nc name
where name is the name of the current run. This decodes the file name.ncepFNL.grib (fi-
nal analysis at 1o resolution), placed in the folder Data/ncepFNL-grib, and creates the file
name.ncepFNL.nc in the folder Data/ncepFNL-nc for subsequent execution of SETDBS.
• Script-ncep1-to-nc name
where name is the name of the current run. This decodes the file name.ncep1.grib (reanalysis
1), placed in the folder Data/ncep1-grib, and creates the file name.ncep1.nc in the folder
Data/ncep1-nc for subsequent execution of SETDBS.
NOTE: This script can be used for ncep1 files already downloaded in grib-format. Currently
ncep1 files are available only in nc-format and in this case Script-ncep1-cat-nc name should
be used.
• Script-ncep2-to-nc name
where name is the name of the current run. This decodes the file name.ncep2.grib (reanalysis
2), placed in the folder Data/ncep2-grib, and creates the file name.ncep2.nc in the folder
Data/ncep2-nc for subsequent execution of SETDBS.
NOTE1: This script can be used for ncep2 files already downloaded in grib-format. Currently
ncep2 files are available only in nc-format and in this case Script-ncep2-cat-nc name should
be used.
NOTE2: Variable files must be named with standard NCEP names, i.e. hgt.nc, rhum.nc,
omega.grib, uwnd.nc, vwnd.nc, etc.
FALL3D-7.2 MANUAL 25

• Script-ncep1-cat-nc name
where name is the name of the current run. This merges the files variable1.nc, variable2.nc,
etc. (reanalysis 1), placed in the folder Data/ncep1-nc, and creates the file name.ncep1.nc
in the folder Data/ncep1-nc for subsequent execution of SETDBS.
• Script-ncep2-cat-nc name
where name is the name of the current run. This merges the files variable1.nc, variable2.nc,
etc. (reanalysis 2), placed in the folder Data/ncep2-nc, and creates the file name.ncep2.nc
in the folder Data/ncep2-nc for subsequent execution of SETDBS.
• Script-era40-to-nc name
where name is the name of the current run. This decodes the file name.era40.grib (ECMWF
ERA-40), placed in the folder Data/era40-grib, and creates the file name.era40.nc in the
folder Data/era40-nc for subsequent execution of SETDBS.
• Script-eraIn-to-nc name
where name is the name of the current run. This decodes the file name.eraIn.grib (ECMWF
ERA-Interim), placed in the folder Data/eraIn-grib, and creates the file name.eraIn.nc in
the folder Data/eraIn-nc for subsequent execution of SETDBS.
• Script-eta-to-nc name
where name is the name of the current run. This decodes the file name.eta.grib (ETA model
output), placed in the folder Data/eta-grib, and creates the file name.eta.nc in the folder
Data/eta-nc for subsequent execution of SETDBS.
• Script-arpa-to-nc name
where name is the name of the current run. This decodes the file name.arpa.grib (ARPA-SIM
model output), placed in the folder Data/arpa-grib, and creates the file name.arpa.nc in the
folder Data/arpa-nc for subsequent execution of SETDBS.
NOTE: If an alias for the scripts has not been previously set, the user must run wgrib/wgrib2
and GRIB2NC from the Scripts folder and provide the full path for the grib file.
NOTE2: For some of the scripts it may be necessary to modify some variables depending on
the specific problem.
3. Run the SETDBS utility program to generate the file name.dbs.nc in the folder Runs/name:
• Script-SetDbs name model
where name is the name of the current run and model is one of the following options:

• gfs05deg/gfs1deg (for global model forecasts)

• ncepFNL/ncep1/ncep2/era40/eraIn (for reanalyses)
• wrf/eta/arpa (for mesoscale models)
• profile/calmet62 (for other options)

4. Run the SETTGSD utility program to generate the file name.tgsd in the folder Runs/name:
• Script-SetTgsd name
where name is the name of the current run. Alternatively, the TGSD file can be created by the user
directly.

5. Run the SETSRC utility program to generate the files name.src and name.grn in the folder Runs/name:
• Script-SetSrc name
where name is the name of the current run.

6. Run FALL3D-7.2 in either serial or parallel versions:

• Script-Fall3d ser name
FALL3D-7.2 MANUAL 26

to run FALL3D-7.2 serial, where name is the name of the current run;
• Script-Fall3d par name ncpu ngroup
to run FALL3D-7.2 parallel. Note that, in general, this script has to be edited and modifyed
depending on each particular queuing system.

7. Run the FALL3D2GMT utility program to postprocess results:

• Script-Fall3d2GMT name
where name is the name of the current run. Note that this makes use of GMT and the convert
utilities (from ImageMagic), not included in the distribution.
FALL3D-7.2 MANUAL 27

Appendices
Appendix A. Governing equations and parameterizations
In FALL3D-7.2 it is assumed that the main factors controlling atmospheric transport of ash are wind
advection, turbulent diffusion, and gravitational settling of particles. This assumption does not hold in
the proximal region that can be extended for large eruptions (having high eruption columns and large
mass eruption rates), where eruption clouds can spread at the NBL as a gravity current. A simple
analytical model describing this effect can be used in FALL3D-7.2.
Neglecting particle-particle interaction effects (collisions, aggregation, etc.), the Eulerian form of the
continuity equation written in a generalized coordinate system (X, Y, Z) is (Byun and Schere, 2006;
Costa et al., 2006):

∂C ∂C ∂C ∂C ∂Vsj
+ VX + VY + (VZ − Vsj ) = −C∇ · V + C
∂t ∂X ∂Y ∂Z ∂Z
      (1)
∂ ∂C/ρ∗ ∂ ∂C/ρ∗ ∂ ∂C/ρ∗
+ ρ∗ KX + ρ∗ KY + ρ∗ KZ + S∗
∂X ∂X ∂Y ∂Y ∂Z ∂Z

where C is the transformed concentration, V = (VX , VY , VZ ) is the transformed wind speed, KX , KY and
KZ are the diagonal terms of the transformed eddy diffusivity tensor, ρ∗ is the transformed atmospheric
density, and S∗ is the transformed source term. FALL3D-7.2 solves Eq. (1) for each particle class j using
a curvilinear terrain-following coordinate system (X = mx, Y = my, z → Z), where m is the map
scale factor and Z = z − h(x, y), with h(x, y) denoting the topographic elevation, and (x, y, z) are the
Cartesian coordinates. The scaling factors for this particular transformation are given in Table 8 (Byun
and Schere, 2006). The generic particle class j is defined by a triplet of values characterizing each particle
(dp , ρp , Fp ), that are, respectively, diameter, density, and a shape factor. For dp we use the equivalent
diameter d, which is the diameter of a sphere of equivalent volume. For the shape factor Fp we choose the
sphericity ψ, which is the ratio of the surface area of a sphere with diameter d to the surface area of the
particle. In our approximation, each triplet (d, ρp , ψ) is sufficient to define the settling velocity. Effect of
Earth’s curvature are considered when the lat-lon coordinate system is used through the Jacobian of the
transformation.

Parameter Scaling
Coordinates X = mx; Y = my; Z = z − h(x, y)
Horizontal Velocities VX = mvx ; VY = mvy
h  i
Vertical velocity (VZ − VSj ) = J −1 (vz − vsj ) − m vx ∂h ∂h
∂x + vy ∂y
Diffusion Coefficients KX = Kx ; KY = Ky ; KZ ' Kz J −2
Concentration C = cJ/m2
Density ρ∗ = ρJ/m2
Source Term S∗ = SJ/m2

Table 8: Scaling factors for a terrain-following coordinate system (x = mX, y = mY, , z → Z). (x, y, z)
are the Cartesian coordinates, m the map scale factor (for the UTM coordinate system m = 1) and J is
the determinant of the Jacobian of the coordinate system transformation.

Eddy Diffusivity Tensor

In FALL3D-7.2 only the diagonal components of the Eddy Diffusivity Tensor, i.e. the vertical Kz and
the horizontal Kh = Kx = Ky components, are considered.
The available choices for describing the vertical component Kz are:
1. Option CONSTANT, i.e. Kz = constant, where the constant value is assigned by the user;
FALL3D-7.2 MANUAL 28

2. Option SIMILARITY. In this case, inside the Atmospheric Boundary Layer (ABL), FALL3D-7.2
evaluates Kz as:
  −1
 z hz
 κu∗ z 1 − 1 + 9.2 h/L ≥ 0 stable


h Lh
Kz =  1/2 (2)
 
 κu∗ z 1 − z hz
 1 − 13 h/L ≤ 0 unstable
h Lh

where κ is the von Karman constant (κ = 0.4), u∗ is the wind friction velocity, h is the ABL height,
and L is the Monin-Obukhov length (see Costa et al., 2006). The expression above comes from an
extension of the Monin-Obukhov similarity theory to the entire ABL (Ulke, 2000). On the other
hand, above the ABL (z/h > 1), Kz is considered as a function of the local vertical wind gradient, a
characteristic length scale lc , and a stability function Fc which depends on the Richardson number
Ri:
∂U
Kz = lc2 Fc (Ri) (3)
∂z
q
where U = yx2 + u2y . For lc and Fc , FALL3D-7.2 adopts the relationship used by the CAM-3.0
model (Collins et al., 2004):
 −1
1 1
lc = + (4)
κz λc
1


 stable (Ri > 0)
Fc (Ri) = 1 + 10Ri(1 + 8Ri) (5)
 √

1 − 18Ri unstable (Ri < 0)
where λc is the so-called asymptotic length scale (λc ≈ 30m).

3. Option SURFACE LAYER. This option can be used to estimated the vertical turbulent diffusivity Kz ,
inside the atmospheric surface layer, in terms of friction velocity u∗ and the Monin-Obukhov length
L:
κzu∗
Kz = (6)
φh
where κ is the von Karman constant (κ = 0.4), z is the distance from the ground and φh is the
atmospheric stability function for temperature (see Jacobson, 1999):
 z
 P rt + βh L
z/L > 0 stable
φh = z −1/2 (7)
P rt 1 − γh L z/L < 0 unstable

P rt z/L = 0 neutral

with βh = 7.8, γh = 11.6 and P rt ≈ 0.95 (Högström, 1988).

The available choices for describing the horizontal component Kh = Kx = Ky are:

1. Option CONSTANT, i.e. Kh = constant, where the constant value is assigned by the user;
2. Option RAMS. In this case, a large eddy parameterization as the one used by the RAMS model
(Pielke et al., 1992) can be used for evaluating Kh :
 v " 
u
u ∂vx 2 2  2 #
∂v y ∂vx ∂vy
Kh = P rt max km ; (CS ∆)2 t + +2 +  (8)
∂y ∂x ∂x ∂y

√
where P rt is the turbulent Prandtl number (typically P rt ≈ 1), km = 0.075∆4/3 , ∆ = ∆x∆y,
∆x and ∆y are the horizontal grid spacings, and CS is a constant ranging from 0.135 to 0.32.
FALL3D-7.2 MANUAL 29

3. Option CMAQ. In this case, the horizontal diffusion is evaluated as in the CMAQ model (Byun and
Schere, 2006):
1 1 1
= + (9)
Kh Kht Khn
where: s 2  2
∂vx ∂vy ∂vy ∂vx
Kht = α2 ∆x∆y − + + (10)
∂x ∂y ∂x ∂y
 
∆xf ∆yf
Khn = Khf (11)
∆x∆y
where the numeric constant α = 0.28 and the values of Khf and ∆xf = ∆yf depend on the
algorithm. Using this parameterization, for a large grid size the effect of the transportive dispersion
is minimized, whereas for a small grid size the numerical diffusion term is reduced (Byun and
Schere, 2006). Thanks to the heuristic relationship (9), the smaller of Kht and Khn dominates. In
our case we set Khf = 8000 m2 s−1 for ∆xf = ∆yf = 4 km and a minimum value for Kh equal to
km = 0.075∆4/3 was imposed.

Settling velocity models

There are several semi-empirical parameterizations for the particle settling velocity vs if one assumes that
particles settle down at their terminal velocity:
s
4g (ρp − ρa ) d
vs = (12)
3Cd ρa

where ρa and ρp denote air and particle density, respectively, d is the particle equivalent diameter, g is the
gravity acceleration, and Cd is the drag coefficient. Cd depends on the Reynolds number, Re = dvs /νa
(νa = µa /ρa is the kinematic viscosity of air, and µa the dynamic viscosity). In FALL3D-7.2 different
options are possible for estimating settling velocity, such as:
1. ARASTOOPOUR model (Arastoopour et al., 1982):
 24
 (1 + 0.15Re0.687 ) Re ≤ 988.947
Cd = Re (13)

0.44 Re > 988.947

valid for spherical particles only.

2. GANSER model (Ganser, 1993):

24 n 0.6567
o 0.4305K2
Cd = 1 + 0.1118 (Re K1 K2 ) + (14)
ReK1 3305
1+
Re K1 K2

where K1 = 3/[(dn /d) + 2ψ −0.5 ], K2 = 101.8148(−Logψ)

0.5743
are two shape factors (dn is the average
between the minimum and the maximum axis, d is the equal volume sphere), and ψ is the particle
sphericity (ψ = 1 for spheres). For calculating the sphericity is practical to use the concepts of
“operational” and “working sphericity”, ψwork introduced by Wadell (1933); Aschenbrenner (1956),
which are based on the determination of the volume and of the three dimensions of a particle
respectively:
(P 2 Q)1/3
ψwork = 12.8 p (15)
1 + P (1 + Q) + 6 1 + P 2 (1 + Q2 )
with P = S/I, Q = I/L, where L is the longest particle dimension, I is the longest dimension
perpendicular to L, and S is the dimension perpendicular to both L and I.
FALL3D-7.2 MANUAL 30

3. WILSON model (Walker et al., 1971; Wilson and Huang, 1979) using the interpolation suggested by
Pfeiffer et al. (2005):

24 −0.828 p
 Re ϕ


 +2 1−ϕ Re ≤ 102
Cd = 1 − Cd |Re=102 (16)
1− (103 − Re) 102 ≤ Re ≤ 103
900


Re ≥ 103

 1

where ϕ = (b + c)/2a is the particle aspect ratio (a ≥ b ≥ c denote the particle semi-axes).
4. DELLINO model (Dellino et al., 2005):
νa
0.5206
vs = 1.2605 Ar ξ 1.6 (17)
d
where Ar = gd3 (ρp − ρa )ρa /µ2a is the Archimedes number, g the gravity acceleration, and ξ is a
particle shape factor (sphericity to circularity ratio). It is recommended to not extrapolate this
option for particle diameters beyond the range used in the experiments by Dellino et al. (2005).
5. DIOGUARDI model (Dioguardi et al., 2017):
 −0.2
0.5134
4.18 −(Re )
4 0.559 Cd,sphere Re Ψ 3D
Cd = (18)
3 Re2
where Ψ3D denotes the 3D sphericity and Cd,sphere is the drag for a sphere calculated according to
(Clift and Gauvin, 1971).
Since for FALL3D-7.2 the primary particle shape factor is the sphericity ψ, for sake of simplicity, ϕ in
(16) and ξ in (17) are calculated by approximating particles as prolate ellipsoids (the same approximation
is used for estimating dn ).

Particle aggregation
For computational reasons ash particle aggregation is assumed to occur within the eruption plume affect-
ing the original TGSD which is modified considering an effective aggregate class and depleting particle
classes finer than the aggregate class itself. A few possibilities are available in FALL3D-7.2 such as

1. NONE option that neglect aggregation;

2. PERCENTAGE option that simply subtracts a constant percentage of ash from each particle class
having diameter smaller than the user-defined aggregate diameter with a user-defined aggregate
density (e.g. Sulpizio et al., 2012);

3. CORNELL option that is similar to the parameterization proposed by Cornell et al. (1983) as used in
Costa et al. (2012);
4. COSTA option based on Costa et al. (2010) model. For computational reasons all the three options
assume that ash aggregation occurs mainly within the eruption column and affect the original
TGSD (described by filename.tgsd) by creating an effective particle distribution (described by
filename.grn). Option COSTA can be used only with source model PLUME.

Source term
FALL3D-7.2 reads the time-dependent source term (mass released per unit time at each grid point) from
an external file. This file can be generated by the SETSRC utility program, choosing among different
options such as:
1. POINT SOURCE that emits mass from a point source only;
FALL3D-7.2 MANUAL 31

2. SUZUKI that describes the eruptive column as a mushroom-like shape Suzuki (1983); Pfeiffer et al.
(2005);
3. HAT that describes the eruptive column as rectangular shape of an assigned thickness;
4. PLUME that uses an eruptive column model based on the Buoyant Plume Theory (based on Folch
et al., 2016).
5. RESUSPENSION that describes resuspension of ash deposited on the ground remobilized by wind.
This option is described more in detail below.

Resuspension of ash
Saltation impact represents the most effective mechanism for resuspension of smaller-size particles in soils
(Shao et al., 1993). When the intensity of wind blowing across a granular soil exceeds a certain threshold,
grain particles begin to saltate. Experiments with sand-sized particles show that the impact of saltating
mid-size grains (larger than about 50 µm) breaks the cohesive forces of smaller particles, enhancing their
suspension when falling back to ground. For this reason, the emission rate (vertical flux of particles),
defined as the mass emitted per unit of area and time, strongly depends on the horizontal (saltation) flux
of larger particles. FALL3D-7.2 uses different emission schemes for ash resuspension by wind (see Folch
et al., 2014):
1. Emission scheme 1 (Westphal et al., 1987) computes the emission rate as:

0 u∗ < u∗t
FV = (19)
10−5 u4∗ u∗ ≥ u∗t
where FV is the vertical flux (in kg m−2 s−1 ), occurring only above a (constant) threshold friction
velocity u∗t . An important limitation of (19) is that the vertical flux does not depend on particle
size or soil moisture. Although very simple, this parameterization can be useful when information
on soil characteristics (e.g. particle sizes and densities, moisture, roughness, etc.) is not available
or poorly constrained.
2. Emission scheme 2 (Marticorena and Bergametti, 1995; Marticorena et al., 1997) computes the
emission rate as: 
 0 u∗ < u∗t (d)
FV (d) = Kρa u∗ 2 2

(20)
u∗ − u∗t (d) u∗ ≥ u∗t (d)
g


where K is a soil texture coefficient equal to K = 5.4 × 10−4 m−1 from experiments, u∗ denotes
the wind friction velocity, and u∗t is the threshold friction velocity given by:
 0.129K
(1.928Re0.092 −1)0.5 0.03 < Re ≤ 10
u∗t = −0.0617(Re−10) (21)
0.129K(1 − 0.0858e ) Re > 10
r  
ρp gd
with K = ρa 1 + ρ0.006
p gd2.5 and Re = 1331 × d1.56 (the lower bound of the fit corresponds to
particles of ≈ 10 µm in size). Please note that inn the relationship above, ρp and ρa are particle
and air densities (expressed in g/cm3 ), g is gravity (in cm/s2 ), d is the particle size (in cm), Re is
the Reynolds number parameterized as a function of the particle size, and u∗t is given in cm/s.
3. Emission scheme 3 (Shao et al., 1993; Shao and Leslie, 1997; Shao and Lu, 2000) computes the
emission rate as:
α(d, ds )
FV (d, ds ) = 2 FH (ds ) (22)
u∗t (d)
where α (units of m s−2 ) is a coefficient of blasting efficiency determined experimentally (Shao and
Leslie, 1997) and FH is the horizontal flux (units of kg m−1 s−1 ) of saltating particles of size ds :
(
0  u∗ < u∗t (ds )
FH (ds ) = ρ u3 u2 (d ) (23)

co ag ∗ 1 − ∗tu2 s u∗ ≥ u∗t (ds )
∗
FALL3D-7.2 MANUAL 32

and co is an empirical dimensionless constant close to 1. The threshold friction velocity u∗t (d) is
given by: s  
ρp gd γ
u∗ts = 0.0123 + (24)
ρa ρa d
where γ is a parameter ranging between 1.65 × 10−4 and 5 × 10−4 kg/s2 (a value of 3 × 10−4 kg/s2
is assumed in FALL3D-7.2).
NOTE: in the current version, simulation of resuspension is possible only in combination with WRF-ARW
meteorological data.

Spreading of the volcanic cloud at the NBL

When the option GRAVITY CURRENT is used an analytical model describing the spreading of the volcanic
cloud at NBL as a gravity current (e.g., Woods and Kienle, 1994; Sparks et al., 1997) is coupled to the
ADS transport (see Costa et al., 2013). In summary this option consists of adding an effective radial
velocity field to the wind field. The radial wind field due to the gravitational spreading of the current is
centered above the vent in the umbrella region and extended up to a radius R in accordance with:
  1/2
2λN q 1
√ for R ≤ Rp and HU − h2 ≤ z ≤ HU + h2

ub (R) = (25)
3π R
for R > Rp or z < HU − h2 or z > HU + h2

ub (R) = 0

where λ is an empirical constant, and N is the frequency of Brunt–Vaisala due to the ambient stratifi-
cation, q is the volumetric flow rate into the umbrella region, HU denotes the level (NBL), and h the
thickness of umbrella region that is assumed to scale with ub as h = ub /(λN ). Within this region variation
of the velocity field with the radial distance r is calculated as:

1 r2
 
3 R
ub (r) = ub (R) 1+ (26)
4 r 3 R2

The radial field is considered negligible at distances larger than a critical radius Rp (Costa et al., 2013).

Wet deposition
As a first approach, wet deposition is assumed below the Planetary Boundary Layer (PBL) only. Using
this approximation only the total rate is necessary to describe wet deposition that is parameterized as
(e.g., Jung and Shao, 2006):
dC
= −L C = −A P B C (27)
dt
where P is the precipitation rate in mm h−1 , A and B are two empirical constants (A = 8.4 · 10−5 and
B = 0.79 respectively). Two critical cut-off size at 1 and 100 µms are assumed.
FALL3D-7.2 MANUAL 33

Appendix B. Example of control input file

This is an example (not from a real case) of control input file. Comments begin with an exclamation
symbol.

!
! --------------------------------------------
!
! EXAMPLE OF FALL3D INPUT FILE
! VERSION: 7.2
!
! --------------------------------------------
!
! --------------------------------------------
! Blocks read by Fall3d and utility programs
! --------------------------------------------
!
TIME UTC
YEAR = 2008
MONTH = 04
DAY = 29
BEGIN METEO DATA (HOURS AFTER 00) = 0
TIME STEP METEO DATA (MIN) = 60
END METEO DATA (HOURS AFTER 00) = 24
ERUPTION START (HOURS AFTER 00) = 0 3.5
ERUPTION END (HOURS AFTER 00) = 7
RUN END (HOURS AFTER 00) = 10
RESTART = NO
!
GRID
COORDINATES = LON-LAT
LON LAT
LONMIN = 14.0
LONMAX = 16.0
LATMIN = 36.5
LATMAX = 38.5
LON VENT = 15.0
LAT VENT = 37.5
VENT HEIGHT (M) = 3000.
NX = 51
NY = 51
ZLAYER (M) FROM 0. TO 8000. INCREMENT 500.
!
! --------------------------------------------
! Block read by SetTGsd utility program
! --------------------------------------------
!
GRANULOMETRY
!
DISTRIBUTION = BIGAUSSIAN
! options: GAUSSIAN/BIGAUSSIAN/WEIBULL/BIWEIBULL
!
NUMBER OF CLASSES = 14
FALL3D-7.2 MANUAL 34

FI MEAN = -2.0 1.5

FI DISP = 1.3 1.0
FI RANGE = -3. 8.
DENSITY RANGE = 1000 2500
SPHERICITY RANGE = 0.9 0.9
MIXING FACTOR = 0.5
!
DISTRIBUTION = BIWEIBULL
NUMBER OF CLASSES = 14
FI SCALE = 1.9 5.2
W SHAPE = 0.63 0.87
FI RANGE = -3. 8.
DENSITY RANGE = 1000 2500
SPHERICITY RANGE = 0.9 0.9
MIXING FACTOR = 0.5
!
! --------------------------------------------
! Blocks read by SetSrc utility program
! --------------------------------------------
!
SOURCE
!
! SOURCE TYPE options: POINT/SUZUKI/HAT/PLUME/RESUSPENSION
!
SOURCE TYPE = plume
!
POINT SOURCE
!
HEIGHT ABOVE VENT (M) = 3000. 6000.
! MASS FLOW RATE (KGS) options : ndt values/ESTIMATE-MASTIN/ESTIMATE-DEGRUYTER/ESTIMATE-WOODHOUSE
MASS FLOW RATE (KGS) = ESTIMATE-WOODHOUSE
!
SUZUKI SOURCE
!
HEIGHT ABOVE VENT (M) = 3000. 6000.
MASS FLOW RATE (KGS) = ESTIMATE-WOODHOUSE
A = 4.
L = 1.
! HAT SOURCE
!
HEIGHT ABOVE VENT (M) = 3000. 6000.
MASS FLOW RATE (KGS) = ESTIMATE-WOODHOUSE
THICKNESS (M) = 2000. 2000.
!
PLUME SOURCE
!
SOLVE PLUME FOR = MFR
MFR SEARCH RANGE = 3.0 7.0
HEIGHT ABOVE VENT (M) = 6000.
MASS FLOW RATE (KGS) = 1d3 1d4
EXIT VELOCIY (MS) = 200.
EXIT TEMPERATURE (K) = 1073.
EXIT WATER FRACTION (%) = 1.
!
FALL3D-7.2 MANUAL 35

WIND COUPLING = YES

AIR MOISTURE = YES
LATENT HEAT = NO
REENTRAINMENT = NO
A S = KAMINSKI-C
A V = TATE
! RESUSPENSION SOURCE
!
MAX RESUSPENSION SIZE (MIC) = 100
DEPOSIT THRESHOLD (KGM2) = 1
MAX INJECTION HEIGHT (M) = 1000
EMISSION SCHEME = WESTPHAL
EMISSION FACTOR = 1.0
THRESHOLD UST = 0.3
MOISTURE CORRECTION = no
!
AGGREGATION
!
AGGREGATION MODEL = Cornell
FI AGGREGATES = 2.
DENSITY AGGREGATES = 350.
VSET FACTOR = 1.0
PERCENTAGE (%) = 20.
FRACTAL EXPONENT = 2.99
!
AEROSOLS
!
SO2 = no
SO2 PERCENTAGE (%) = 1.
!
GRAVITY CURRENT
!
! Gravity current model based on Suzuki and Koyaguchi (2009)
! options : YES/NO
!
GRAVITY CURRENT = no
C FLOW RATE = 1d4
LAMBDA GRAV = 0.2
K ENTRAIN = 0.1
BRUNT VAISALA = 0.02
!
! --------------------------------------------
! Block read by Fall3d
! --------------------------------------------
!
FALL3D
!
! TERMINAL VELOCITY MODEL options : ARASTOOPOUR/GANSER/WILSON/DELLINO/DIOGUARDI
! VERTICAL TURBULENCE MODEL options : CONSTANT/SIMILARITY/SURFACE LAYER
! VERTICAL DIFFUSION COEFFICIENT (M2/S) : value. Only used if VERTICAL TURBULENCE MODEL =
CONSTANT
! HORIZONTAL TURBULENCE MODEL options : CONSTANT/RAMS/CMAQ
! RAMS CS : value. Only used if HORIZONTAL TURBULENCE MODEL = RAMS
! HORIZONTAL DIFFUSION COEFFICIENT (M2/S) : value. Only used if HORIZONTAL TURBULENCE MODEL
FALL3D-7.2 MANUAL 36

= CONSTANT
! WET DEPOSITION options : YES/NO
!
TERMINAL VELOCITY MODEL = ganser
VERTICAL TURBULENCE MODEL = CONSTANT
VERTICAL DIFFUSION COEFFICIENT (M2/S) = 500.
HORIZONTAL TURBULENCE MODEL = CONSTANT
RAMS CS = 0.3
HORIZONTAL DIFFUSION COEFFICIENT (M2/S) = 5000.
WET DEPOSITION = yes
!
OUTPUT
!
POSTPROCESS TIME INTERVAL (HOURS) = 1.
POSTPROCESS 3D VARIABLES = YES
POSTPROCESS CLASSES = YES
TRACK POINTS = YES
!
! ----------------------------------------------
! Blocks read by Postprocess utility program(s)
! ----------------------------------------------
!
POSTPROCESS
!
CROP DOMAIN
LONMIN = 14.0
LONMAX = 16.0
LATMIN = 36.5
LATMAX = 38.5
!
! Meteo
!
MAP TEMPERATURE = yes
MAP VELOCITY = yes
Z CUTS (M) = 1000 5000
!
! Time independent variables
!
MAP TOPOGRAPHY = no
UNITS = M
CONTOUR LEVELS = 1. 100. 500. 1000. 1500. 2000. 2500.
!
! Deposit variables
!
MAP TOTAL LOAD = no
UNITS = KG/M2
CONTOUR LEVELS = 0.1 0.5 1. 5. 10. 50.
!
MAP WET LOAD = no
UNITS = KG/M2
CONTOUR LEVELS = 0.1 0.5 1. 5.
!
MAP CLASS LOAD = no
UNITS = KG/M2
FALL3D-7.2 MANUAL 37

CONTOUR LEVELS = 0.1 0.5 1. 5. 10. 50.

!
MAP CLASS WET = yes
UNITS = KG/M2
CONTOUR LEVELS = 0.1 0.5 1. 5.
!
MAP DEPOSIT THICKNESS = no
DEPOSIT DENSITY (KG/M3) = 1100.0 (Default value 1000.)
UNITS = MM (Possibilities: MM/CM/M)
CONTOUR LEVELS = 0.1 0.5 1. 5. 10.
!
! Ground level variables
!
MAP CONCE GROUND = no
UNITS = G/M3
CONTOUR LEVELS = 0.001 0.01 0.1 1.
!
MAP PMxx GROUND = no
UNITS = G/M3
CONTOUR LEVELS = 1d-5 1d-4 1d-3
!
! Airborne variables
!
MAP COLUMN MASS = no
UNITS = G/M2
CONTOUR LEVELS = 1. 10. 100.
!
MAP COLUMN PMxx = no
UNITS = G/M2
CONTOUR LEVELS = 0.1 1. 10. 100.
!
MAP FLIGHT LEVEL = no
UNITS = G/M3
CONTOUR LEVELS = 0.001 0.01 0.1
!
MAP AOD = no
UNITS = NONE
CONTOUR LEVELS = 0.1 0.25 0.5 0.75 1 2
FALL3D-7.2 MANUAL 38

Appendix C. The NetCDF format

NetCDF (network Common Data Form) is a set of software libraries and machine-independent data
formats that support the creation, access, and sharing of array-oriented scientific data
(http://www.unidata.ucar.edu/software/netcdf/).
FALL3D-7.2 uses the standard NetCDF format for both database input file (name.dbs.nc) and results
output file (name.res.nc). There is a good number of open-source codes to view, analyze or manipulate
NetCDF files, for example:
• ncview and ncdump (http://opendap.org/download/nc clients.html).
• Panoply (http://www.giss.nasa.gov/tools/panoply/).

• GMT (http://gmt.soest.hawaii.edu/)
• GrADS (http://www.iges.org/grads/).
• NCL, the NCAR Command Language (http://www.ncl.ucar.edu/).

• GRASS (http://grass.osgeo.org/)
• QGIS (http://www.qgis.org)
• R (http://www.r-project.org/)
FALL3D-7.2 MANUAL 39

pcoord
pdate
itime1 itime2
nz
z(1) ux(1) uy(1) T(1)
...
z(nz) ux(nz) uy(nz) T(nz)
itime3 itime4
...

Table 9: Format of the meteo data file name.profile.dat for the PROFILE case. Repeat this block for
each meteo time increment.

Appendix D. Format of the meteo profile file (name.profile)

For the profile option, the utility SetDbs needs an ASCII file containing the definition of the vertical wind
and temperature profile and a topography file of the domain in GRD format. In this case wind velocities
are assumed constant on all the domain in a terrain-following coordinate system. The remaining variables
are assumed with the values of the Standard Atmosphere. The format of the profile file (name.profile)
is described in Table 9 and the meaning of the used symbols is the following:
• pcoord: Coordinates where the profile was measured; either as UTM or lon-lat coordinates.
• pdate: Starting time when the profile was measured; the format of the date is yyyymmdd, i.e. year,
month, day.

• itime1: Initial time in sec after the starting time pdate of validity of the meteo data contained in
the following nz layers.
• itime2: Final time in sec after the starting time pdate of validity of the meteo data contained in
the following nz layers.

• nz: Number of the database vertical layers.

• z: Vertical coordinate of the layer (in m a.s.l.).
• ux: wind x-velocity (in m/s).
• uy: wind y-velocity (in m/s).

• T: temperature T -velocity (in o C).

FALL3D-7.2 MANUAL 40

Appendix E. The GRD format

The structure of a GRD format file is described in Table 10 and the meaning of the used symbols is the
following:
• NX : Number of grid points along x-direction.

• NY : Number of grid points along y-direction.

• XO : x-coordinate (UTM in m) of the grid bottom left corner point.
• XF : x-coordinate (UTM in m) of the grid top right corner point.

• YO : y-coordinate (UTM in m) of the grid bottom left corner point.

• YF : y-coordinate (UTM in m) of the grid top right corner point.
• VAL : Value at each grid point. It consists of an array of NX×NY values stored starting from the
bottom-left corner and moving towards right then up towards the top-right corner.

NX NY
XO XF
YO YF
MAX(v) MIN(v)
VAL(i,1) ... ... i=1:NX
... ... ...
VAL(i,j) ... ... i=1:NX
... ... ...
VAL(i,NY) ... ... i=1:NX

Table 10: Format of a GRD file name.grd.

FALL3D-7.2 MANUAL 41

Appendix F. Further reading

The following is a list of publications related to FALL3D:
1. Costa, A., Macedonio, G., Folch, A. (2006). A three-dimensional Eulerian model for transport and
deposition of volcanic ashes. Earth Planet. Sci. Lett., 241 (3-4), 634-647,
doi: 10.1016/j.epsl.2005.11.019.
2. Folch, A., Jorba, O., and Viramonte, J. (2008). Volcanic ash forecast - application to the May 2008
Chaitén eruption, Nat. Hazards Earth Syst. Sci., 8, 927-940.
3. Scollo, S., A. Folch, A. Costa (2008). A parametric and comparative study of different tephra
fallout models, J. Volcanol. Geotherm. Res., 176, 199-211, doi: 10.1016/j.jvolgeores.2008.04.002.
4. Folch A., C. Cavazzoni, A. Costa, G. Macedonio (2008). An automatic procedure to forecast tephra
fallout, J. Volcanol. Geotherm. Res., 177, 767-777, doi: 10.1016/j.jvolgeores.2008.01.046 .
5. Macedonio G., A. Costa, A. Folch (2008). Ash fallout scenarios at Vesuvius: Numerical simu-
lations and implications for hazard assessment, J. Volcanol. Geotherm. Res., 178, 366-377, doi:
10.1016/j.jvolgeores.2008.08.014.
6. Folch A., Costa A., Macedonio G. (2009). FALL3D: A Computational Model for Volcanic Ash
Transport and Deposition, Comput. Geosci., 35, 1334-1342, doi: 10.1016/j.cageo.2008.08.008.
7. Scollo S., Prestifilippo M., Spata G., D’Agostino M., Coltelli M. (2009). Monitoring and forecasting
Etna volcanic plumes, Nat. Hazards Earth Syst. Sci., 9, 1573-1585.
8. Costa A., A. Folch, G. Macedonio (2010). A Model for Wet Aggregation of Ash Particles in Volcanic
Plumes and Clouds: I. Theoretical Formulation, J. Geophys. Res., 115, B09201,
doi:10.1029/2009JB007175.
9. Folch A., A. Costa, A. Durant,G. Macedonio (2010). A Model for Wet Aggregation of Ash Par-
ticles in Volcanic Plumes and Clouds: II. Model Application, J. Geophys. Res., 115, B09202,
doi:10.1029/2009JB007176.
10. Folch A., Sulpizio R. (2010). Evaluating long-range volcanic ash hazard using supercomputing fa-
cilities: application to Somma-Vesuvius (Italy), and consequences for civil aviation over the Central
Mediterranean Area, Bull. Volc., 72 (9), 1039-1059, doi:10.1007/s00445-010-0386-3.
11. Scollo S., A. Folch, M. Coltelli, V.J. Realmuto (2010). Three-dimensional volcanic aerosol dis-
persal: A comparison between Multiangle Imaging Spectroradiometer (MISR) data and numerical
simulations, J. Geophys. Res., 115, D24210, doi:10.1029/2009JD013162.
12. Corradini S., Merucci L., Folch A. (2011). Volcanic Ash Cloud Properties: Comparison Between
MODIS Satellite Retrievals and FALL3D Transport Model, IEEE Geoscience and Remote Sensing
Letters, 8, 248-252, doi:10.1109/LGRS.2010.2064156.
13. Folch A., Costa A., Basart S. (2012). Validation of the FALL3D ash dispersion model using obser-
vations of the 2010 Eyjafjallajokull volcanic ash cloud, Atmos. Environ., 48, 165-183,
doi:10.1016/j.atmosenv.2011.06.072.
14. Scaini C., Folch A., Navarro M. (2012). Tephra hazard assessment at Concepciòn Volcano, Nicaragua,
J. Volcanol. Geotherm. Res., Volumes 219-220, 41-51 doi:10.1016/j.jvolgeores.2012.01.007.
15. Costa A., Folch A., Macedonio G., Giaccio B., Isaia R., Smith V.C. (2012) Quantifying volcanic
ash dispersal and impact from Campanian Ignimbrite super-eruption, Geophys. Res. Lett., 39,
doi:10.1029/2012GL051605.
16. Bonasia R., Costa A., Folch A., Capra L., Macedonio G. (2012). Numerical simulation of tephra
transport and deposition of the 1982 El Chichon eruption, J. Volcanol. Geotherm. Res., Vol.
231-232, 39-49, doi: 10.1016/j.jvolgeores.2012.04.006.
FALL3D-7.2 MANUAL 42

17. Sulpizio, R., Folch, A., Costa, A., Scaini, C., Dellino, P. (2012). Civil aviation hazard assessment of
far-range volcanic ash dispersal from a violent Strombolian eruption scenario at Somma-Vesuvius
volcano, Naples, Italy, Bull. Volcanol., 74, 2205-2218, doi: 10.1007/s00445-012-0656-3.
18. Collini, E., Osores, S., Folch, A., Viramonte, J.G., Villarosa, G., Salmuni, G. (2013). Volcanic ash
forecast during the June 2011 Cordón-Caulle, Natural Hazards, 66(2), 389-412, doi:10.1007/s11069-
012-0492-y.
19. Osores, M.S., Folch, A., Collini, E., Villarosa, G., Durant, A., Pujol, G., Viramonte, J.G. (2013).
Validation of the FALL3D model for the 2008 Chaitén eruption using field, laboratory and satellite
data, Andean Geology, 40 (2): 262-276.

20. Folch, A., Mingari L., Osores, M.S., Collini, E. (2013). Modeling volcanic ash resuspension. Appli-
cation to the 14-18 October 2011 outbreak episode in Central Patagonia, Argentina, Nat. Hazards
Earth Syst. Sci., 14, 119-133, doi:10.5194/nhess-14-119-2014.
21. Costa A., Folch A., Macedonio G. (2013) Density-driven transport in the umbrella region of vol-
canic clouds: Implications for tephra dispersion models, Geophys. Res. Lett., Vol. 40, 1-5, doi:
10.1002/grl.50942
22. Costa A., Smith V.C., Macedonio G., Matthews N. (2014) The magnitude and impact of the
Youngest Toba Tuff super-eruption, Front. Earth Sci., Vol. 2, 16, doi: 10.3389/feart.2014.00016
23. Selva J., Costa A., Sandri L., Macedonio G., Marzocchi W. (2014) Probabilistic short-term volcanic
hazard in phases of unrest: a case study for tephra fallout, J. Geophys. Res., Vol. 119 (12), 1-22,
doi: 10.1002/2014JB011252
24. Bonasia R., Scaini C., Capra L., Nathenson M., Arana-Salinas L., Siebe C., Folch A. (2014), Long-
range hazard assessment of volcanic ash dispersal for a Plinian eruptive scenario at Popocatépetl vol-
cano (Mexico): implications on civil aviation, Bull. Volcanol., Vol. 76 (1), 789, doi: 10.1007/s00445-
013-0789-z

25. Parra R. Bernard B., b, Narváez D., Le Pennec J.L., Hasselle N, Folch A. (2016) Eruption Source
Parameters for forecasting ash dispersion and deposition from vulcanian eruptions at Tungurahua
volcano: Insights from field data from the July 2013 eruption, J. Volcanol. Geotherm. Res., Vol.
309, 1-13, doi: 10.1016/j.jvolgeores.2015.11.001
26. Martı́ A., Folch A., Costa A., Engwell A. (2016) Reconstructing the plinian and co-ignimbrite
sources of large volcanic eruptions: a novel approach for the Campanian Ignimbrite, Nature Sci.
Rep., 6, 21220, 1-11; doi:10.1038/srep21220.
27. Sandri L., Costa A., Selva J., Tonini R., Macedonio G., Folch A., Sulpizio R. (2016) Beyond eruptive
scenarios: assessing tephra fallout hazard from Neapolitan volcanoes, Nature Sci. Rep., 6: 24271,
1-13, doi:10.1038/srep24271.

28. Folch A., Costa A., Macedonio G. (2016) FPLUME-1.0: An integral volcanic plume model account-
ing for ash aggregation, Geosci. Model Dev., 9, 431-450, doi:10.5194/gmd-9-431-2016
29. Macedonio G., Costa A., Folch A. (2016) Uncertainties in volcanic plume modeling: A parametric
study using FPLUME, J. Volcanol. Geotherm. Res., 326, 92-102, doi:10.1016/j.jvolgeores.2016.03.016
FALL3D-7.2 MANUAL 43

References
Arastoopour, H., Wang, C. H., and Weil, S. A.: Particle-particle interaction force in a diluite gas-solid
system, Chem. Eng. Sci., 37, 1379–1386, 1982.
Aschenbrenner, B. C.: A new method of expressing particle sphericity, J. Sediment. Petrol., 26, 15–31,
1956.

Bonadonna, C. and Phillips, J. C.: Sedimentation from strong volcanic plumes, J. Geophys. Res., 108,
2340, doi:10.1029/2002JB002034, 2003.
Byun, D. and Schere, K. L.: Review of the governing equations, computational algorithms, and other
components of the Models-3 Community Multiscale Air Quality (CMAQ) modeling system, Applied
Mechanics Reviews, 59, 51–77, doi:10.1115/1.2128636, 2006.

Clift, R. and Gauvin, W. H.: Motion of entrained particles in gas stream, Can. J. Chem. Eng., 49,
439–448, doi:10.1002/cjce.5450490403, 1971.
Collins, W. D., Rasch, P. J., Boville, B. A., Hack, J. J., McCaa, J. R., Williamson, D. L., Kiehl, J. T., and
Briegleb, B.: Description of the NCAR Community Atmosphere Model (CAM 3.0), Technical Report
NCAR/TN-464+STR, National Center for Atmospheric Research, Boulder, Colorado, 2004.

Cornell, W., Carey, S., and Sigurdsson, H.: Computer simulation and transport of the Campanian Y-5
ash, J. Volcanol. Geotherm. Res., 17, 89–109, doi:10.1016/0377-0273(83)90063-X, 1983.
Costa, A., Macedonio, G., and Folch, A.: A three-dimensional Eulerian model for transport and deposition
of volcanic ashes, Earth Planet. Sci. Lett., 241, 634–647, 2006.

Costa, A., Folch, A., and Macedonio, G.: A model for wet aggregation of ash particles in volcanic plumes
and clouds: I. Theoretical formulation, J. Geophys. Res., 115, doi:10.1029/2009JB007175, 2010.
Costa, A., Folch, A., Macedonio, G., Giaccio, B., Isaia, R., and Smith, V. C.: Quantifying volcanic
ash dispersal and impact from Campanian Ignimbrite super-eruption, Geophys. Res. Lett., 39, doi:
10.1029/2012GL051605, 2012.

Costa, A., Folch, A., and Macedonio, G.: Density-driven transport in the umbrella region of volcanic
clouds: Implications for tephra dispersion models, Geophys. Res. Lett., 40, 1–5, doi:10.1002/grl.50942,
2013.
Costa, A., Pioli, L., and Bonadonna, C.: Assessing tephra total grain-size distribution: Insights from
field data analysis, Earth Planet. Sci. Lett., 443, 90–107, doi:10.1016/j.epsl.2016.02.040, 2016.

Costa, A., Pioli, L., and Bonadonna, C.: Corrigendum to “Assessing tephra total grain-size distribution:
Insights from field data analysis” [Earth Planet. Sci. Lett. 443 (2016) 90–107], Earth Planet. Sci. Lett.,
465, 205–209, doi:10.1016/j.epsl.2017.03.003, 2017.
Degruyter, W. and Bonadonna, C.: Improving on mass flow rate estimates of volcanic eruptions, Geophys.
Res. Lett., 39, L16 308, doi:10.1029/2012GL052566, 2012.
Dellino, P., Mele, D., Bonasia, R., Braia, L., and La Volpe, R.: The analysis of the influence of pumice
shape on its terminal velocity, Geophys. Res. Lett., 32, doi:10.1029/2005GL023954, 2005.
Dioguardi, F., Mele, D., Dellino, P., and Dürig, T.: The terminal velocity of volcanic particles with
shape obtained from 3D X-ray microtomography, J. Volcanol. Geotherm. Res., 329, 41–53, doi:
10.1016/j.jvolgeores.2016.11.013, 2017.
Fecan, F., Marticorena, B., and Bergametti, G.: Parametrization of the increase of the aeolian erosion
threshold wind friction velocity due to soil moisture for arid and semi-arid areas, Ann. Geophys. Italy,
17, 149–157, 1999.
FALL3D-7.2 MANUAL 44

Folch, A., Costa, A., and Macedonio, G.: FALL3D: A computational model for transport and deposition
of volcanic ash, Comput. Geosci., 35, 1334–1342, doi:10.1016/j.cageo.2008.08.008, 2009.
Folch, A., Mingari, L., Osores, M. S., and Collini, E.: Modeling volcanic ash resuspension - application to
the 14-18 October 2011 outbreak episode in Central Patagonia, Argentina, Nat. Hazards Earth Syst.
Sci., 14, 119–133, doi:10.5194/nhess-14-119-2014, 2014.

Folch, A., Costa, A., and Macedonio, G.: FPLUME-1.0: An integral volcanic plume model accounting
for ash aggregation, Geosci. Model Dev., 9, 431–450, doi:10.5194/gmd-9-431-2016, 2016.
Ganser, G. H.: A rational approach to drag prediction of spherical and nonspherical particles, Powder
Technol., 77, 143–152, doi:10.1016/0032-5910(93)80051-B, 1993.

Högström, U.: Non-dimensional wind and temperature profiles in the atmospheric surface layer: A re-
evaluation, Bound-Lay. Meteorol., 42, 55–78, doi:10.1007/BF00119875, 1988.
Jacobson, M. Z.: Fundamentals of atmospheric modelling, Cambridge University Press, New York, 1st
edn. edn., 1999.

Jung, E. and Shao, Y.: An intercomparison of four wet deposition schemes used in dust transport
modeling, Global and Planetary Change, 52, 248–260, 2006.
Kaminski, E., Tait, S., , and Carazzo, G.: Turbulent entrainment in jets with arbitrary buoyancy, J.
Fluid Mech., 526, 361–376, doi:10.1017/S0022112004003209, 2005.
Marticorena, B. and Bergametti, G.: Modeling the atmospheric dust cycle: 1. Design of a soil-derived
dust emission scheme, J. Geophys. Res., 100, 16 415–16 430, doi:10.1029/95JD00690, 1995.
Marticorena, B., Bergametti, G., Aumont, B., Callot, Y., N’Doumé, C., and Legrand, M.: Modeling
the atmospheric dust cycle 2. Simulation of Saharan dust sources, J. Geophys. Res., 102, 4387–4404,
doi:10.1029/96JD02964, 1997.

Mastin, L. G., Guffanti, M., Servranckx, R., Webley, P., Barsotti, S., Dean, K., Durant, A., Ewert, J. W.,
Neri, A., Rose, W. I., Schneider, D., Siebert, L., Stunder, B., Swanson, G., Tupper, A., Volentik, A.,
and Waythomas, C. F.: A multidisciplinary effort to assign realistic source parameters to models of
volcanic ash-cloud transport and dispersion during eruptions, J. Volcanol. Geotherm. Res., 186, 10–21,
doi:10.1016/j.jvolgeores.2009.01.008, 2009.

Pfeiffer, T., Costa, A., and Macedonio, G.: A model for the numerical simulation of tephra fall deposits,
J. Volcanol. Geotherm. Res., 140, 273–294, doi:10.1016/j.jvolgeores.2004.09.001, 2005.
Pielke, R. A., Cotton, W. R., Walko, R. L., Tremback, C. J., Nicholls, M. E., Moran, M. D., Wesley,
D. A., Lee, T. J., and Copeland, J. H.: A comprehensive meteorological modeling system-RAMS,
Meteor. Atmos. Phys., 49, 69–91, 1992.

Rouse, H., Yih, C. S., and Humphreys, H. W.: Gravitational convection from a boundary source, Tellus,
4, 201–210, 1952.
Scire, J. S., Robe, F. R., Fernau, M. E., and Yamartino, R. J.: A User’s Guide for the CALMET
Meteorological Model, Tech. Rep. Version 5, Earth Tech, Inc., 196 Baker Avenue, Concord, MA 01742,
2000.

Shao, Y. and Leslie, L. M.: Wind erosion prediction over the Australian continent, J. Geophys. Res., 102,
30,091–30,105, 1997.
Shao, Y. and Lu, H.: A simple expression for wind erosion threshold friction velocity, J. Geophys. Res.,
105, 22 437–22 443, doi:10.1029/2000JD900304, 2000.

Shao, Y., Raupach, M. R., and Findlater, P. A.: Effect of saltation bombardment on the entrainment of
dust by wind, J. Geophys. Res., 98, 12,719–12,726, doi:10.1029/93JD00396, 1993.
FALL3D-7.2 MANUAL 45

Sparks, R. S. J., Bursik, M. I., Carey, S. N., Gilbert, J. S., Glaze, L. S., Sigurdsson, H., and Woods,
A. W.: Volcanic Plumes, John Wiley & Sons Ltd., Chichester, U.K., 1997.
Sulpizio, R., Folch, A., Costa, A., Scaini, C., and Dellino, P.: Hazard assessment of far-range volcanic ash
dispersal from a violent Strombolian eruption at Somma-Vesuvius volcano, Naples, Italy: Implications
on civil aviation, Bull. Volcanol., 74, 2205–2218, doi:10.1007/s00445-012-0656-3, 2012.

Suzuki, T.: A theoretical model for dispersion of tephra, in: Arc Volcanism: Physics and Tectonics, edited
by Shimozuru, D. and Yokoyama, I., pp. 93–113, Terra Scientific Publishing Company (TERRAPUB),
Tokyo, 1983.
Suzuki, Y. and Koyaguchi, T.: A three-dimensional numerical simulation of spreading umbrella clouds,
J. Geophys. Res., 114, B03 209, doi:10.1029/2007JB005369, 2009.
Tate, P. M.: The rise and dilution of buoyant jets and their behaviour in an inter-
nal wave field, Phd thesis, University of New South Wales, School of Mathematics, URL
http://trove.nla.gov.au/version/19798635, last access: 16 September 2015, 2002.
Ulke, A. G.: New turbulent parameterization for a dispersion model in atmospheric boundary layer,
Atmos. Environ., 34, 1029–1042, 2000.
Wadell, H.: Sphericity and roundness of rock particles, J. Geol., 41, 310–331, 1933.
Walker, G. P. L., Wilson, L., and Bowell, E. L. G.: Explosive volcanic eruptions I. Rate of fall of
pyroclasts, Geophys. J. Roy. Astron. Soc., 22, 377–383, 1971.

Westphal, D. L., Toon, O. B., and Carlson, T. N.: A two-dimensional numerical investigation of the
dynamics and microphysics of Saharan dust storms, J. Geophys. Res., 92, 3027–3049, 1987.
Wilson, L. and Huang, T. C.: The influence of shape on the atmospheric settling velocity of volcanic ash
particles, Earth Planet. Sci. Lett., 44, 311–324, doi:10.1016/0012-821X(79)90179-1, 1979.

Woodhouse, M. J., Hogg, A. J., Phillips, J. C., and Sparks, R. S. J.: Interaction between volcanic plumes
and wind during the 2010 Eyjafjallajökull eruption, Iceland, J. Geophys. Res. Solid Earth, 118, 92–109,
doi:10.1029/2012JB009592, 2013.
Woods, A. W. and Kienle, J.: The dynamics and thermodynamics of volcanic clouds: Theory and
observations from the April 15 and April 21, 1990 eruptions of Redoubt Volcano, Alaska, J. Volcanol.
Geotherm. Res., 62, 273–299, 1994.