Building a climate risk assessment requires data, and lots of it. Climate projections, historical hazard records, water stress indicators, elevation models, and land cover classifications all feed into a comprehensive physical risk analysis. The challenge is finding reliable, global climate database sources that cover what you need.
This guide organizes the best free climate risk data sources by hazard type, the way professional assessments actually use them. Whether you’re a risk manager building in-house capabilities, an ESG analyst sourcing data for TCFD disclosure, or a developer integrating climate data into applications, you’ll find the datasets you need. For a complete walkthrough of how these data sources fit into an end-to-end workflow, see our climate risk analytics guide.
What Makes a Good Global Climate Database?
Before diving into specific datasets, consider what makes climate risk data useful for assessments:
Resolution: Climate data ranges from ~25km (global models) to 30m (terrain data). Higher resolution captures local variation but requires more processing. For portfolio screening, 25km is often sufficient. For site-specific analysis, you need finer resolution context data.
Temporal coverage: Historical baselines typically use 1980-2010 or 1950-2014. Future projections extend to 2050, 2080, or 2100 depending on the dataset. TCFD recommends assessing at least two time horizons.
Scenario coverage: Modern assessments use Shared Socioeconomic Pathways (SSPs) from IPCC AR6. At minimum, you need SSP2-4.5 (moderate emissions) and SSP5-8.5 (high emissions) for scenario comparison.
Update frequency: Climate projections are static (based on CMIP6 model runs), but observational data like fire detections and flood events update regularly.
Climate Projection Datasets
Climate projections form the foundation of any forward-looking risk assessment. These datasets provide temperature, precipitation, and wind projections under different emissions scenarios.
NASA NEX-GDDP-CMIP6 is the workhorse dataset for climate risk assessments. It provides bias-corrected, statistically downscaled projections from CMIP6 climate models at ~25km resolution. Variables include daily mean, minimum, and maximum temperature, precipitation, and surface wind speed. SSP2-4.5 and SSP5-8.5 scenarios are available from 2015 to 2100.
CMIP6 Archive (ESGF) provides raw climate model outputs from 35+ global models. Use this when you need multi-model ensembles or variables not available in NEX-GDDP. The trade-off is complexity: raw model outputs require bias correction and often have varying resolutions.
Copernicus Climate Data Store hosts ERA5 reanalysis data, which provides historical climate conditions from 1950 to present at ~30km resolution. ERA5 is essential for establishing baseline conditions against which future projections are compared.
CORDEX provides regionally downscaled projections at 12-50km resolution for specific domains (Africa, Europe, North America, etc.). Use CORDEX when you need higher resolution for a specific region.
Temperature Hazard Data
Temperature hazards include heat waves, cold stress, and long-term warming. These drive impacts on labor productivity, energy demand, and infrastructure.
For future projections, NEX-GDDP-CMIP6 provides the core temperature variables (tas, tasmax, tasmin). Heat wave assessments typically count days exceeding a threshold above the historical mean maximum temperature.
ERA5 Reanalysis provides historical temperature baselines. You need these to calculate anomalies and establish what “normal” looks like for a given location.
Berkeley Earth provides long-term temperature records extending back to 1850, useful for understanding historical trends and validating model outputs.
Precipitation and Drought Data
Precipitation hazards include drought, extreme rainfall, and changes in seasonal patterns. These affect water availability, agriculture, and flood risk.
NEX-GDDP-CMIP6 provides precipitation projections, but converting raw precipitation to drought indices requires additional processing.
CHIRPS (Climate Hazards Group InfraRed Precipitation with Station data) provides high-resolution (5km) historical rainfall data, particularly valuable for tropical and subtropical regions where station data is sparse.
GPCC (Global Precipitation Climatology Centre) provides global precipitation climatology based on rain gauge data. Use this for historical baseline validation.
SPEI Global Drought Monitor provides pre-computed Standardized Precipitation-Evapotranspiration Index values, a drought indicator that accounts for both precipitation deficits and temperature-driven evaporation.
Water Stress and Flood Data
Water stress represents long-term supply-demand imbalances, while flood risk captures acute inundation events. These require different data sources.
WRI Aqueduct 4.0 is the leading global water stress dataset. It provides basin-level projections for water stress, groundwater depletion, and drought risk under multiple scenarios. Projections are available for 2030, 2050, and 2080. This is the dataset physical climate risk assessments use for water stress hazards.
Global Flood Database maps historical flood events from satellite observations, providing event-level data on inundation extent and duration.
JRC Global Surface Water tracks changes in surface water extent from 1984 to present at 30m resolution. Use this to understand water body dynamics and historical flood patterns.
GloFAS (Global Flood Awareness System) provides river discharge forecasts and historical simulations at ~10km resolution.
Sea Level Rise Data
Sea level rise affects coastal properties, infrastructure, and entire communities. Assessment requires both sea level projections and high-resolution coastal elevation data.
IPCC AR6 Sea Level Rise Projections provide the authoritative global projections by scenario. For SSP5-8.5, median sea level rise is projected at 0.40m by 2050 and continues accelerating through 2100. These are global values; local sea level can differ due to land subsidence or uplift.
NASA Sea Level Change provides observed sea level trends from satellite altimetry, essential for validating model projections against actual measurements.
CoastalDEM from Climate Central provides coastal elevation data corrected for vegetation bias, improving flood exposure estimates in densely vegetated coastal areas.
Fire and Compound Hazard Data
Wildfire risk depends on temperature, humidity, wind, and vegetation. Fire Weather Index calculations require multiple variables, though not all are reliably available from climate models.
NASA FIRMS (Fire Information for Resource Management System) provides near-real-time active fire detection at 375m to 1km resolution. Use this for historical fire occurrence patterns and current monitoring.
EFFIS/GlobFire from the EU Joint Research Centre provides historical fire perimeters as polygon data, useful for analyzing burned area extent and fire spread patterns.
Canadian Forest Fire Weather Index System documentation explains the standard methodology for fire weather calculation. The full FWI requires temperature, relative humidity, wind speed, and precipitation.
Terrain and Context Data
Terrain data provides critical context for hazard calculations. Elevation determines flood susceptibility and sea level rise exposure. Slope affects landslide risk. Land cover influences wildfire behavior.
NASA SRTM (Shuttle Radar Topography Mission) provides global elevation data at 30m resolution. This is the standard dataset for terrain classification in climate risk assessments.
ESA WorldCover provides global land cover classification at 10m resolution, distinguishing between forests, cropland, urban areas, water bodies, and other cover types. Land cover determines wildfire susceptibility and influences local climate conditions.
ALOS World 3D from JAXA provides an alternative global elevation dataset at 30m resolution, useful for cross-validation or areas where SRTM has data gaps.
Historical Disaster Databases
Historical disaster data helps validate risk models and understand past impacts. These databases complement forward-looking climate projections.
EM-DAT (Emergency Events Database) maintained by CRED/UCLouvain records disasters worldwide since 1900, including economic losses, fatalities, and affected populations. This is the standard source for historical disaster impact analysis.
NOAA Storm Events Database provides detailed records of storms and severe weather in the United States, including county-level impacts and narrative descriptions.
NOAA Billion-Dollar Disasters tracks US weather and climate disasters exceeding $1 billion in damages, providing long-term economic loss trends.
Aggregated Indices and Portals
Sometimes you need country-level risk rankings rather than raw data. These indices aggregate multiple factors into composite scores.
Germanwatch Climate Risk Index ranks countries by climate-related losses and fatalities, updated annually. Useful for country-level risk comparisons and understanding historical vulnerability.
ND-GAIN Index (Notre Dame Global Adaptation Initiative) measures country vulnerability and readiness to adapt to climate change. It combines exposure, sensitivity, and adaptive capacity indicators.
World Bank Climate Change Knowledge Portal provides country-level climate risk profiles combining historical data, future projections, and vulnerability indicators.
Choosing the Right Data for Your Assessment
The right data depends on your use case:
For TCFD disclosure: You need scenario-based projections (at least two scenarios) across multiple time horizons. TCFD recommendations call for physical risk assessment using climate scenarios. NEX-GDDP-CMIP6 combined with WRI Aqueduct covers most hazards.
For portfolio screening: 25km resolution is typically sufficient for identifying high-risk locations across hundreds or thousands of assets. Start with composite indices, then drill down to site-level data for flagged locations.
For site-specific analysis: Combine climate projections with high-resolution terrain (SRTM) and land cover (WorldCover) data. Consider local factors that global models miss.
Platforms like Continuuiti aggregate these datasets so you don’t have to process raw climate model outputs yourself. A single API call returns multi-hazard assessments across scenarios and time horizons, with the data processing handled automatically.
Frequently Asked Questions
What resolution do I need for TCFD climate risk disclosure?
For TCFD disclosure, 25km resolution climate projections are generally acceptable for portfolio-level analysis. TCFD focuses on scenario-based assessment across time horizons, not engineering-grade precision. However, for material assets, you may need higher resolution terrain and flood data.
Can I use free climate data for commercial risk assessments?
Yes. Most datasets listed here are freely available for commercial use. NASA, NOAA, ESA, and IPCC data are public domain or open access. Check specific license terms for datasets from research institutions like WRI Aqueduct, but most permit commercial applications.
What is the difference between CMIP5 and CMIP6?
CMIP6 is the latest generation of climate model projections, released for IPCC AR6. It includes more models, higher resolution, and updated emissions scenarios (SSPs instead of RCPs). For new assessments, use CMIP6-based datasets like NEX-GDDP-CMIP6.
How do I handle uncertainty in climate projections?
Use multiple scenarios (at least SSP2-4.5 and SSP5-8.5) to capture emissions pathway uncertainty. For model uncertainty, multi-model ensembles are ideal but computationally intensive. At minimum, understand that single-model projections represent one plausible future, not a prediction.
How often is climate projection data updated?
Climate projection datasets like NEX-GDDP-CMIP6 are static, based on model runs conducted for IPCC assessment cycles. New CMIP generations appear roughly every 7-10 years. Observational data like fire detections (FIRMS) and flood events update daily to monthly.
Climate risk assessment requires assembling data from multiple sources, each covering different hazards and time scales. The datasets in this guide provide the building blocks for comprehensive physical risk analysis. For commercial platforms that handle this complexity for you, the trade-off is convenience versus cost and customization.
