Reference Damage Curves Tutorial#

This tutorial demonstrates how to run a damage analysis using reference damage curves in RA2CE. Reference curves (such as the Huizinga et al. dataset) are widely used for rapid assessments of hazard-related damage, as they provide generic hazard–damage relationships without requiring local calibration.

RA2CE provides several built-in options for vulnerability curves:

Global → Huizinga curves (HZ analysis name)
Europe → OSDamage functions (OSD analysis name)

These curves determine how hazard intensity (e.g., water depth) is translated into expected damage fractions.

Important Notes#

The Huizinga curves are the global default option and are widely used in international flood damage assessments. For more details, see publication by Huizinga et al. (2017): Global flood depth-damage functions: Methodology and the database with guidelines
The OSDamage functions provide Europe-specific depth–damage relations. For more details, see publication by van Ginkel et al. (2021): Flood risk assessment of the European road network.

In this example, we will use flood depth maps as hazard input and apply the Huizinga reference curves to estimate direct damages to road infrastructure.

Step 1: Define project paths#

We first define the root project folder and its subdirectories:

[ ]:

from pathlib import Path

root_dir = Path("data", 'damage_reference_curves')

static_path = root_dir.joinpath("static")
hazard_path = static_path.joinpath("hazard")
network_path = static_path.joinpath("network")
output_path = root_dir.joinpath("output")

Step 2: Configure the road network and hazard#

The network is downloaded from OpenStreetMap (OSM), clipped to a region polygon (polygon.geojson). We specify which road types should be included in the analysis.

[ ]:

from ra2ce.network.network_config_data.enums.road_type_enum import RoadTypeEnum
from ra2ce.network.network_config_data.enums.source_enum import SourceEnum
from ra2ce.network.network_config_data.enums.network_type_enum import NetworkTypeEnum
from ra2ce.network.network_config_data.network_config_data import NetworkSection

network_section = NetworkSection(
    network_type=NetworkTypeEnum.DRIVE,
    source=SourceEnum.OSM_DOWNLOAD,
    polygon=static_path.joinpath("polygon.geojson"),
    save_gpkg=True,
    road_types=[
        RoadTypeEnum.SECONDARY,
        RoadTypeEnum.SECONDARY_LINK,
        RoadTypeEnum.PRIMARY,
        RoadTypeEnum.PRIMARY_LINK,
        RoadTypeEnum.TRUNK,
        RoadTypeEnum.MOTORWAY,
        RoadTypeEnum.MOTORWAY_LINK,
    ],
)

We provide hazard input in the form of GeoTIFF raster files (e.g., flood depth maps). RA2CE will overlay these rasters with the road network to compute hazard intensities for each asset.

[ ]:

from ra2ce.network.network_config_data.enums.aggregate_wl_enum import AggregateWlEnum
from ra2ce.network.network_config_data.network_config_data import HazardSection

hazard_section = HazardSection(
    hazard_map=[Path(file) for file in hazard_path.glob("*.tif")],
    aggregate_wl=AggregateWlEnum.MEAN,  # mean water depth used in analysis
    hazard_crs="EPSG:4326",  # ensure hazard map is in EPSG:4326 projection
)

We combine the network and hazard information into a single configuration object.

[ ]:

from ra2ce.network.network_config_data.network_config_data import NetworkConfigData

network_config_data = NetworkConfigData(
    root_path=root_dir,
    static_path=static_path,
    network=network_section,
    hazard=hazard_section
)
network_config_data.network.save_gpkg = True

Step 3: Define the damage analysis#

Here, we specify that RA2CE should perform a damage analysis using the Huizinga reference damage curves (HZ) with the class AnalysisSectionDamages and the attribute damage_curve set to DamageCurveEnum.HZ

The event type can be set to EVENT if damages are to be calculated for the hazard maps only (example below), or to RETURN_PERIOD if the analysis should estimate risk over a specified return period (see tutorial ).

[ ]:

from ra2ce.analysis.damages.damages import AnalysisSectionDamages
from ra2ce.analysis.analysis_config_data.enums.analysis_damages_enum import AnalysisDamagesEnum
from ra2ce.analysis.analysis_config_data.enums.event_type_enum import EventTypeEnum
from ra2ce.analysis.analysis_config_data.enums.damage_curve_enum import DamageCurveEnum
from ra2ce.analysis.analysis_config_data.analysis_config_data import AnalysisConfigData

damages_analysis = [AnalysisSectionDamages(
    name='damages_reference_curve_Huizinga',
    analysis=AnalysisDamagesEnum.DAMAGES,
    event_type=EventTypeEnum.EVENT,
    damage_curve=DamageCurveEnum.HZ,  # use Huizinga reference curve
    save_csv=True,
    save_gpkg=True
)]

analysis_config_data = AnalysisConfigData(
    analyses=damages_analysis,
    output_path=output_path,
)

Step 4: Run the analysis#

Finally, we run the analysis using the RA2CE handler.

[ ]:

from ra2ce.ra2ce_handler import Ra2ceHandler

Ra2ceHandler.run_with_config_data(network_config_data, analysis_config_data)

Output#

The results of the Huizinga damage analysis are provided in two GeoPackage (GPKG) files:

damages_reference_curve_Huizinga_link_based.gpkg: damage estimates per network link (from node to node).
damages_reference_curve_Huizinga_segment.gpkg: damage estimates per segment of 100m along the network.

Key attributes of interest in the output (expressed in currency) include:

dam_EV1_HZ : estimated damage for the first flood map (Huizinga method).
dam_EV2_HZ : estimated damage for the second flood map (Huizinga method).

You can open these files in GIS software (QGIS, ArcGIS) or load them in Python using GeoPandas for further analysis:

[ ]:

import geopandas as gpd
output_path = root_dir / "output" / 'damages'
output_path.exists()
link_based = gpd.read_file(output_path / "damages_reference_curve_Huizinga_link_based.gpkg")
segment_based = gpd.read_file(output_path / "damages_reference_curve_Huizinga_segmented.gpkg")

# Inspect the first rows
print(link_based.head())
print(segment_based.head())

You can open the results in GIS software to visualize which road segments are most affected by the hazard.

Note#

Reference damage curves provide generalized estimates of vulnerability. For more locally calibrated studies, consider using manual damage curves.