Configuration

Further configuration details can be added here when the functionality becomes available.

Here we will take a look at the conf.py file to see how to configure the hazardmap settings and data sources:

Hazard options

These flags (True/False) determine which hazard inputs will be used to calculate the overall hazard map at the end.

SEISMIC = True
FLOODING = True
VOLCANIC = True
LANDSLIDE = False
EXPOSURE = True

Custom vulnerability curve

Setting this to true will make the TOMRAP tool look for a csv file that defines vulnerability rating curves by building type as a csv file.

An example custom vulnerability curve is located in the repository here:

Example vulnerability curve for Tanzania buildings

Vulnerability curves work in place of the list of building weights found at the bottom of the config file. (See further down this document.) Instead of using the same multiplier for every building type in all cases, the user can specifiy a hazard likelihood in this config file (hazard_intensity) and then the weighting for that specific likelihood is read from the vulnerability curve supplied. TOMRAP does not require the exact hazard intensity specified to match the values supplied in the vulnerability curve. For example, if you specify a hazard intensity in the config file here that falls between two points in the csv file, TOMRAP will choose the closest matching value in the csv file. The vuln_curve_file is relative to the DATADIR directory.

CUSTOM_VULN_CURVE = True

hazard_intensity = 2.0
vuln_curve_file = "vuln_curve_example.csv"

See also the section on Calculating vulnerability curves

Invert coordinates

This is a bit of a workaround if your flood data triggers an index error, this can be due to swapping the x and y coordinates. The Nepa data requires this to be true but not Tanzania

invert_flood_tiff = False

Data locations and types

DATADIR = "../datadir/Tanzania/"
# The one above is not found in the data folder - DV added on Friday - was in the other folder
#config.exposure_file = config.DATADIR + "TZA_buildings_exposure_20200731.dbf" #contains location id and positions  - present
exposure_file = DATADIR + "TZA_buildings_exposure_20200731.dbf" #contains location id and positions  - present
exposure_breakdown_file = DATADIR + "TZA_buildings_exposure_breakdown_20200731.dbf"

Volcano data specifications

volcP is a list of volcano code names used to calculate Pyroclastic flow hazards.

volcL is a list of volcano code names used to calculate Lahar hazards.

volcfile is the shapefile containing the volcano point data (referenced by the code names). You would have to inspect this file first to retrieve the code names used.

volcnames is a list of all the volcanos used in the analysis.

volcP = ["kyejo", "meru"]
volcL = ["lengai", "ngozi", "rungwe"]
volcfile = DATADIR + "World_Volcanoes_Smithsonian_Institution_GVP.shp"
volcnames = ["Lengai, Ol Doinyo", "Meru", "Ngozi", "Rungwe", "Kyejo"]

Flood data specifications

flood_ratio refers to the return period of the flood data you have. So if you have a series of flood TIFFs separated by 100 year return period intervals, you would specificy “100” here.

The filenames must match the structure below, i.e.

FLOODTYPE_1inXXX.tif, e.g PU_1in100.tif (that’s a one not an L, i.e. “one in a hundred”)

PU_1in100.tif    # One in 100 year pluvial flood map
PU_1in200.tif    # One in 200 year pluvial flood map
PU_1in300.tif

FD_1in100.tif    # One in 100 year fluvial (defended) flood map
FD_1in200.tif
FD_1in300.tif
..and so on

Should be specified in the config file as:

# 1 in 100, 1 in 200 in the file
floodratio = 100 # selects from different flood tifs
floodtypes = ["FD", "FU", "P"]  #selects from different flood tifs

floodtypes is a list of codes for the type of flood in the tiff.

P - Pluvial

FD - Fluvial - defended

FU - Fluvial - undefended

Earthquake data supplementary info

At present, the only parameter needed is the earthquake file DBF file.

eearthquake_file = DATADIR + "hazard_map_mean_tanzania.dbf"   (yes that is two Es in eearthquake...)

Output plot files

The figure_prefix option simply appends some text to the start of every output figure to help with organising your files later. It can be left as an empty string to append no text: ‘’ if you prefer.

The plot_types is a python list of the various plot types supported by TOMRAP. You can specify as many or as few as you like.

figure_prefix = "output_"
plot_types = ["ear", "plu", "flu", "tep", "lahar", "pgaindx", "P", "FU", "lah", "pyr", "equ", "flood", "volc", "hmap"]

The plot type codes are explained in the table below:

Plot code	Explanation
ear	Earthquake
plu	Pluvial
flu	Fluvial
tep	Tephra
lahar	Lahars
P	—
FU	—
pgaindx	Peak Ground Acceleration
pyr	Pyroclastic flows
equ	Combined Earthquake Hazard
flood	Combined Flood Hazard
volc	Combined Volcanic Hazard
hmap	Combined All Hazards

Pluvial Flooding is that caused by rainfall alone, Fluvial flooding is that which is caused by the overflowing of a water body (a river, for example.)

hmap is the “final” combined hazard map, though you may have use cases where you want to produce the component hazard risks as well.

Building types

building_type_tz

The building type codes are used to calculate risk based on building type, either from the values specified in the config file, or the vulnerability curve csv file supplied. The codes below should match the ones in the dbf files.

building_type_tz = ['CR/LFM/HBET:1,3',
                    'CR/LFM/HBET:4,7',
                    'CR/LFM/HBET:8,20',
                    'CR/LFINF+DNO/HBET:1,3',
                    'CR/LFINF+DNO/HBET:4,7',
                    'CR/LFINF+DNO/HBET:8,20',
                    'S',
                    'MUR+CB99/HBET:1,3',
                    'MUR+CB99/HBET:4,7',
                    'W',
                    'MATO/LN',
                    'MUR+ADO/HBET:1,3',
                    'MUR+CL99',
                    'MUR+STRUB',
                     'W+WWD']

Manual weightings

The set of weightings at the end of the config file is used to set the building weights for each type.

The order of the list for each type matches that in the building_type_tz list above. So each item in the list is an element-wise weight for the building type. Therefore, the lists below have to be the same length as the building_type_tz list above. Note that each hazard type has a separate weighting list. Currently only earthquakes can be set to have a vulnerability curve defined in a csv file, as of May 2022, you must set the other hazard weighting types using this list method below. You may set weights to be the same by using the syntax tz_weight_fluvial = tz_weight_pluvial for example, to set fluvial and pluvial weightings to use the same weights if you wanted to.

# Building type weightings
tz_weight_pluvial = [0.32, 0.2, 0.12, 0.4, 0.25, 0.15, 0.09, 0.4, 0.25, 0.8, 0.56, 0.56, 0.56, 0.56, 0.56]
tz_weight_fluvial = tz_weight_pluvial
tz_weight_tephra = [0.3, 0.15, 0.09, 0.4, 0.2, 0.12, 0.09, 0.5, 0.25, 0.2, 0.6, 0.6, 0.6, 0.6, 0.6]
tz_weight_lahar = [0.06, 0.1, 0.06, 0.6, 0.3, 0.18, 0.3, 0.4, 0.2, 1, 1, 1, 1, 1, 1]
tz_weight_pyro = [0.56, 0.63, 0.7, 0.64, 0.72, 0.8, 0.9, 0.72, 0.81, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8]
tz_weight_earthquake = [0.12, 0.32, 0.16, 0.18, 0.48, 0.24, 0.2, 0.09, 0.24, 0.09, 0.3, 0.3, 0.3, 0.3, 0.3]