README.md

---
license: mit
task_categories:
- image-segmentation
- image-classification
- other
language:
- en
tags:
- geospatial
- hazards
- climate
- risk-assessment
- disaster-management
- flood
- earthquake
- volcano
- tsunami
- tropical-cyclone
- landslide
- drought
- storm-surge
size_categories:
- 1M<n<10M
configs:
- config_name: default
  data_files: "**/*"
---

# Global Multi-Hazard Risk Assessment Dataset

## Dataset Description

This dataset contains comprehensive global multi-hazard risk assessment data covering 9 different natural hazard types. The data is sourced from the UNEP Grid and represents state-of-the-art hazard modeling for disaster risk assessment and climate change adaptation planning.

### Dataset Summary

- **Total Size**: ~50GB of geospatial data
- **Coverage**: Global
- **Resolution**: Variable (1km to 11km depending on hazard type)
- **Coordinate System**: WGS84 (EPSG:4326)
- **Format**: GeoTIFF (.tif) and Shapefile (.shp)
- **Time Periods**: Current climate (1979-2016) and future climate scenarios (2017-2100)

## Hazard Types

### 1. **Flood Hazards** 🌊
- **Files**: 18 TIFF files
- **Return Periods**: 2, 5, 10, 25, 50, 100 years
- **Climate Scenarios**: 
  - Existing climate (1979-2016)
  - SSP1 Lower bound (RCP2.6)
  - SSP5 Upper bound (RCP8.5)
- **Model**: Continuum hydrological model with Manning equation hydraulic modeling
- **Resolution**: ~1km
- **Total Size**: ~50GB

### 2. **Earthquake Hazards** 🌍
- **Files**: 1 TIFF file
- **Parameter**: Peak Ground Acceleration (PGA) for 475-year return period
- **Model**: Global seismic hazard assessment
- **Resolution**: ~3 minutes (~5.5km at equator)

### 3. **Volcano Hazards** 🌋
- **Files**: Shapefile format (Volcano.shp + associated files)
- **Content**: Global volcano locations and attributes
- **Format**: Vector data (can be converted to raster)
- **Source**: Global volcano database

### 4. **Tropical Cyclone Hazards** 🌀
- **Files**: 6 TIFF files
- **Parameter**: Wind velocity
- **Return Periods**: 25, 50, 100 years
- **Climate Scenarios**: Current climate and climate change (RCP8.5)
- **Resolution**: 6 minutes (~11.1km at equator)
- **Total Size**: ~50MB

### 5. **Tsunami Hazards** 🌊
- **Files**: 1 GeoPackage file
- **Parameter**: Tsunami run-up height on coastline
- **Return Period**: 475 years
- **Coverage**: Coastal areas with tsunami hazard
- **Format**: Vector data

### 6. **Landslide Hazards** 🏔️
- **Files**: 4 TIFF files
- **Types**: 
  - Earthquake-triggered landslides
  - Precipitation-triggered landslides
- **Climate Scenarios**: Existing climate and SSP5 RCP8.5
- **Parameter**: Susceptibility classes

### 7. **Drought Hazards** ☀️
- **Status**: Metadata available, files pending
- **Coverage**: Global drought risk assessment

### 8. **Storm Surge Hazards** 🌊
- **Status**: Metadata available, files pending
- **Coverage**: Coastal storm surge risk

## Data Structure

```
dataset/
├── flood/                    # Flood hazard data
│   ├── metadata.json
│   └── *.tif                 # 18 flood hazard TIFF files
├── earthquake/               # Earthquake hazard data
│   ├── metadata.json
│   └── *.tif                 # PGA data
├── volcano/                  # Volcano hazard data
│   ├── metadata.json
│   ├── Volcano.shp          # Shapefile
│   ├── Volcano.dbf          # Database file
│   ├── Volcano.prj          # Projection file
│   └── Volcano.shx          # Index file
├── tropical_cyclone/         # Tropical cyclone data
│   ├── metadata.json
│   └── *.tif                 # 6 wind velocity files
├── tsunami/                  # Tsunami hazard data
│   ├── metadata.json
│   └── *.gpkg                # GeoPackage file
├── landslide/                # Landslide hazard data
│   ├── metadata.json
│   └── *.tif                 # 4 susceptibility files
├── drought/                  # Drought hazard data
│   └── metadata.json
└── storm_surge/              # Storm surge hazard data
    └── metadata.json
```

## Usage

### Loading the Dataset

```python
import rasterio
import geopandas as gpd
import os

# Load a flood hazard TIFF
flood_data = rasterio.open('flood/flood_hazard_100_years_-_existing_climate.tif')
print(f"Shape: {flood_data.shape}")
print(f"CRS: {flood_data.crs}")
print(f"Bounds: {flood_data.bounds}")

# Load volcano shapefile
volcano_data = gpd.read_file('volcano/Volcano.shp')
print(f"Number of volcanoes: {len(volcano_data)}")
print(f"Columns: {volcano_data.columns.tolist()}")
```

### Converting Shapefiles to Raster

```python
import subprocess

# Convert volcano shapefile to optimized TIFF with 1km resolution
subprocess.run([
    'gdal_rasterize',
    '-burn', '1',
    '-tr', '0.009', '0.009',  # 1km resolution
    '-a_nodata', '0',
    '-ot', 'Byte',
    '-co', 'COMPRESS=LZW',
    '-co', 'TILED=YES',
    'volcano/Volcano.shp',
    'volcano_optimized_1km.tif'
])
```

## Data Processing

### Optimization for Large Files

The flood hazard files are large (~3-4GB each). For efficient processing:

1. **Use NoData values** instead of storing zeros
2. **Apply compression** (LZW or DEFLATE)
3. **Use tiled format** for better performance
4. **Consider data type optimization** (Byte vs Float32)

### Coordinate System

All data uses WGS84 (EPSG:4326) coordinate system:
- **Longitude**: -180° to +180°
- **Latitude**: -90° to +90°
- **Units**: Decimal degrees

## Applications

This dataset is suitable for:

- **Risk Assessment**: Multi-hazard risk modeling
- **Climate Adaptation**: Future scenario planning
- **Disaster Management**: Emergency response planning
- **Research**: Climate change impact studies
- **Policy Making**: Infrastructure planning
- **Insurance**: Risk pricing and underwriting

## Citation

If you use this dataset, please cite:

```bibtex
@dataset{global_multi_hazard_2024,
  title={Global Multi-Hazard Risk Assessment Dataset},
  author={UNEP Grid and INGENIAR Risk},
  year={2024},
  url={https://hazards-data.unepgrid.ch/},
  note={Multi-hazard disaster risk model of infrastructure and buildings at the global level}
}
```

## References

- **Flood Data**: Silvestro et al. (2013, 2015) - Continuum hydrological model
- **Climate Data**: W5E5 Global Meteorological dataset, ISIMIP3b
- **Background Paper**: [Multi-hazard Disaster Risk Model](https://giri.unepgrid.ch/sites/default/files/2023-11/2.4-INGENIAR-CDRI-Background-Report-Risk-model.pdf)
- **Source**: [UNEP Grid](https://unepgrid.ch/en)

## License

This dataset is released under the MIT License. See the license file for details.

## Contact

For questions about this dataset, please contact:
- **UNEP Grid**: https://unepgrid.ch/en
- **INGENIAR Risk**: https://www.ingeniar-risk.com/en

## Updates

- **v1.0** (2024): Initial release with 9 hazard types
- **Future**: Additional hazard types and higher resolution data planned

---

*This dataset represents a comprehensive global multi-hazard risk assessment suitable for research, policy-making, and disaster risk management applications.*