Build Model

Using the python API.

Here we are going to build a model using the HydroMT-FIAT python API.

# Imports
import tempfile
from pathlib import Path

from hydromt._utils import log

from hydromt_fiat import FIATModel
from hydromt_fiat.data import fetch_data

Lets make sure that there is a directory to build to.

Also ensure we have some data to build with, hydromt_fiat comes with a small to function to fetch a small dataset to build with.

# Use the built-in tempfile package
temp_dir = tempfile.TemporaryDirectory(delete=False)
p = Path(temp_dir.name)

## Get the build data
data_dir = fetch_data("build-data")

Downloading file 'build-data.tar.gz' from 'doi:10.5281/zenodo.17495611/build-data.tar.gz' to '/home/runner/.cache/hydromt_fiat'.
Untarring contents of '/home/runner/.cache/hydromt_fiat/build-data.tar.gz' to '/home/runner/.cache/hydromt_fiat/build-data'

Now lets setup the Model.

# Create a model in write mode.
log._setuplog()
model = FIATModel(root=p, mode="w+", data_libs=Path(data_dir, "data_catalog.yml"))

2025-11-03 12:10:33,515 - hydromt - log - INFO - HydroMT version: 1.2.0
2025-11-03 12:10:33,516 - hydromt.data_catalog.data_catalog - data_catalog - INFO - Parsing data catalog from /home/runner/.cache/hydromt_fiat/build-data/data_catalog.yml
2025-11-03 12:10:34,074 - hydromt.model.root - root - INFO - setting root to /tmp/tmps8oderk8
2025-11-03 12:10:34,085 - hydromt.model.model - model - INFO - Initializing fiat model from hydromt_fiat (v1.0.0.dev0).

Next step is to update the configurations and add a region for the model to know to collect data in.

Luckily a region vector file is incorporated in the small build data directory.

# Add some of the configurations
model.setup_config(**{"model.model_type": "geom", "output.path": "output"})

# Add a region to build with
model.setup_region(Path(data_dir, "region.geojson"))

2025-11-03 12:10:34,090 - hydromt.hydromt_fiat.fiat - fiat - INFO - Setting config entries from user input
2025-11-03 12:10:34,091 - hydromt.hydromt_fiat.fiat - fiat - INFO - Setting region from '/home/runner/.cache/hydromt_fiat/build-data/region.geojson'

Lets verify the presence of our region, which is crucial for building our hazard and exposure data.

model.region

	geometry
0	POLYGON ((4.38362 51.97582, 4.38125 51.97865, ...

All seems in order, therefore we can move on to setting up the FIAT model data.

Let’s start with the vulnerability, as this data is crucial for the exposure data.

# Call the setup method of the component
model.vulnerability.setup(
    vulnerability_fname="vulnerability_curves",
    vulnerability_linking_fname="vulnerability_curves_linking",
    unit="m",
    continent="europe",  # A keyword argument to select data from the
    # vulnerability_fname dataset
)

2025-11-03 12:10:34,117 - hydromt.hydromt_fiat.components.vulnerability - vulnerability - INFO - Setting up the vulnerability curves
2025-11-03 12:10:34,119 - hydromt.data_catalog.drivers.dataframe.pandas_driver - pandas_driver - INFO - Reading using pandas driver from /home/runner/.cache/hydromt_fiat/build-data/vulnerability/jrc_damage_functions.csv
2025-11-03 12:10:34,130 - hydromt.data_catalog.drivers.dataframe.pandas_driver - pandas_driver - INFO - Reading using pandas driver from /home/runner/.cache/hydromt_fiat/build-data/vulnerability/jrc_damage_functions_linking.csv

This should result in damage curve data in our vulnerability component, lets see.

model.vulnerability.data.curves

	water depth	rs1	cm1	in1	in2	un1
0	0.00	0.000	0.000	0.000	0.000	0.0000
1	0.01	0.005	0.003	0.003	0.005	0.0049
2	0.02	0.010	0.006	0.006	0.010	0.0098
3	0.03	0.015	0.009	0.009	0.015	0.0146
4	0.04	0.020	0.012	0.012	0.020	0.0195
...	...	...	...	...	...	...
996	9.96	1.000	1.000	1.000	1.000	1.0000
997	9.97	1.000	1.000	1.000	1.000	1.0000
998	9.98	1.000	1.000	1.000	1.000	1.0000
999	9.99	1.000	1.000	1.000	1.000	1.0000
1000	10.00	1.000	1.000	1.000	1.000	1.0000

1001 rows × 6 columns

Simply done.

Let’s continue with the hazard data.

Hazard data is easily setup from a raster-based datasource.

# Call the setup method from the hazard component
model.hazard.setup(
    hazard_fnames="flood_event",
    hazard_type="water depth",
)

2025-11-03 12:10:34,259 - hydromt.hydromt_fiat.components.hazard - hazard - INFO - Setting up hazard raster data
2025-11-03 12:10:34,346 - hydromt.hydromt_fiat.workflows.hazard - hazard - INFO - Added water depth hazard map: flood_event
2025-11-03 12:10:34,346 - hydromt.hydromt_fiat.workflows.utils - utils - WARNING - No known grid provided to reproject to, defaulting to first specified grid for transform and extent

Let’s take a look a the current state of the hazard data after settting it up with the setup method.

model.hazard.data

<xarray.Dataset> Size: 4kB
Dimensions:      (y: 34, x: 25)
Coordinates:
  * y            (y) float64 272B 4.458e+05 4.456e+05 ... 4.426e+05 4.424e+05
  * x            (x) float64 200B 8.525e+04 8.535e+04 ... 8.755e+04 8.765e+04
    spatial_ref  int64 8B 0
Data variables:
    flood_event  (y, x) float32 3kB dask.array<chunksize=(34, 25), meta=np.ndarray>
Attributes:
    type:      water depth
    analysis:  event

xarray.Dataset

• Dimensions:
  • y: 34
  • x: 25
• Coordinates: (3)
  • y
    (y)
    float64
    4.458e+05 4.456e+05 ... 4.424e+05

    axis :

    Y

    long_name :

    Northing

    standard_name :

    projection_y_coordinate

    units :

    metre

    array([445750., 445650., 445550., 445450., 445350., 445250., 445150., 445050.,
           444950., 444850., 444750., 444650., 444550., 444450., 444350., 444250.,
           444150., 444050., 443950., 443850., 443750., 443650., 443550., 443450.,
           443350., 443250., 443150., 443050., 442950., 442850., 442750., 442650.,
           442550., 442450.])

  • x
    (x)
    float64
    8.525e+04 8.535e+04 ... 8.765e+04

    axis :

    X

    long_name :

    Easting

    standard_name :

    projection_x_coordinate

    units :

    metre

    array([85250., 85350., 85450., 85550., 85650., 85750., 85850., 85950., 86050.,
           86150., 86250., 86350., 86450., 86550., 86650., 86750., 86850., 86950.,
           87050., 87150., 87250., 87350., 87450., 87550., 87650.])

  • spatial_ref
    ()
    int64
    0

    crs_wkt :

    PROJCRS["Amersfoort / RD New",BASEGEOGCRS["Amersfoort",DATUM["Amersfoort",ELLIPSOID["Bessel 1841",6377397.155,299.1528128,LENGTHUNIT["metre",1]]],PRIMEM["Greenwich",0,ANGLEUNIT["degree",0.0174532925199433]],ID["EPSG",4289]],CONVERSION["unnamed",METHOD["Oblique Stereographic",ID["EPSG",9809]],PARAMETER["Latitude of natural origin",52.1561605555556,ANGLEUNIT["degree",0.0174532925199433],ID["EPSG",8801]],PARAMETER["Longitude of natural origin",5.38763888888889,ANGLEUNIT["degree",0.0174532925199433],ID["EPSG",8802]],PARAMETER["Scale factor at natural origin",0.9999079,SCALEUNIT["unity",1],ID["EPSG",8805]],PARAMETER["False easting",155000,LENGTHUNIT["metre",1],ID["EPSG",8806]],PARAMETER["False northing",463000,LENGTHUNIT["metre",1],ID["EPSG",8807]]],CS[Cartesian,2],AXIS["easting",east,ORDER[1],LENGTHUNIT["metre",1]],AXIS["northing",north,ORDER[2],LENGTHUNIT["metre",1]],ID["EPSG",28992]]

    spatial_ref :

    PROJCRS["Amersfoort / RD New",BASEGEOGCRS["Amersfoort",DATUM["Amersfoort",ELLIPSOID["Bessel 1841",6377397.155,299.1528128,LENGTHUNIT["metre",1]]],PRIMEM["Greenwich",0,ANGLEUNIT["degree",0.0174532925199433]],ID["EPSG",4289]],CONVERSION["unnamed",METHOD["Oblique Stereographic",ID["EPSG",9809]],PARAMETER["Latitude of natural origin",52.1561605555556,ANGLEUNIT["degree",0.0174532925199433],ID["EPSG",8801]],PARAMETER["Longitude of natural origin",5.38763888888889,ANGLEUNIT["degree",0.0174532925199433],ID["EPSG",8802]],PARAMETER["Scale factor at natural origin",0.9999079,SCALEUNIT["unity",1],ID["EPSG",8805]],PARAMETER["False easting",155000,LENGTHUNIT["metre",1],ID["EPSG",8806]],PARAMETER["False northing",463000,LENGTHUNIT["metre",1],ID["EPSG",8807]]],CS[Cartesian,2],AXIS["easting",east,ORDER[1],LENGTHUNIT["metre",1]],AXIS["northing",north,ORDER[2],LENGTHUNIT["metre",1]],ID["EPSG",28992]]

    GeoTransform :

    85200.0 100.0 0.0 445800.0 0.0 -100.0

    x_dim :

    x

    y_dim :

    y

    array(0)

• Data variables: (1)
  • flood_event
    (y, x)
    float32
    dask.array<chunksize=(34, 25), meta=np.ndarray>

    AREA_OR_POINT :

    Area

    _FillValue :

    nan

    source_file :

    event.tif

    grid_mapping :

    spatial_ref

    name :

    flood_event

    Array Chunk Bytes 3.32 kiB 3.32 kiB Shape (34, 25) (34, 25) Dask graph 1 chunks in 5 graph layers Data type float32 numpy.ndarray

• Attributes: (2)

  type :

  water depth

  analysis :

  event