{"success":true,"data":{"author":"World Bank / GFDRR and CSIRO; COG mirror by Portolan Mirrors","bucketAccessType":"public","bucketListable":true,"bucketName":"us-west-2.opendata.source.coop","bucketPrefix":"portolan-mirrors/wb-fwi","category":"Climate & Weather","cloudProvider":"aws","createdAt":"2026-11-18T17:37:22.675Z","description":"Global wildfire hazard grids from the World Bank, computed by CSIRO for the ThinkHazard! wildfire layer: for each 0.5 degree cell, the yearly maximum Canadian Fire Weather Index (FWI) expected once in 2, 10 and 30 years, derived from daily FWI in NASA's Global Fire Weather Database (GFWED) for 1981 to 2016. One collection per return period (fwi-rp2/, fwi-rp10/, fwi-rp30/), each a single Cloud Optimized GeoTIFF such as csiro_wf_max_fwi_rp2.tif with embedded statistics, converted from the World Bank GeoTIFFs with pixel values unchanged; a STAC catalog and an agent guide describe the quirks, for example that 0 marks masked oceans and deserts as well as cells with no fire weather. Used for screening wildfire hazard in disaster risk assessments, infrastructure planning and climate risk studies. Read with rasterio or GDAL. Licensed CC BY 4.0 as listed in the World Bank Data Catalog; attribute the World Bank and CSIRO.","displayTitle":"World Bank Global Wildfire Hazard: Fire Weather Index (FWI) Return Levels, 2, 10 and 30 year, 0.5 degree (COG)","fuseMountBrowsable":true,"id":"8a9e99eb-e137-53dc-8d8e-a0bb020e32a5","license":"CC-BY-4.0","licenseEvidenceUrl":"https://datacatalog.worldbank.org/search/dataset/0066912/global-wildfire-hazard","objectCountEstimate":3,"probeStatus":"ok","region":"us-west-2","schemaCard":{"accessHints":[{"language":"python-anonymous","snippet":"import boto3\nfrom botocore import UNSIGNED\nfrom botocore.client import Config\ns3 = boto3.client('s3', region_name='us-west-2', config=Config(signature_version=UNSIGNED))\ns3.list_objects_v2(Bucket='us-west-2.opendata.source.coop', Prefix='portolan-mirrors/wb-fwi/', Delimiter='/')"},{"language":"python-local","snippet":"open('/mnt/archive/fwi-rp10/AGENTS.md', 'rb')"}],"caveats":["Files are binary — use a domain-specific parser (e.g. metpy / pynexrad / pydicom) rather than text I/O."],"format":"binary","generatedAt":"2026-11-18T17:39:26.816Z","layout":{"example":"portolan-mirrors/wb-fwi/fwi-rp10/AGENTS.md","pattern":"{prefix}/{seg1}/{seg2}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"# Wildfire hazard: FWI, 10-year return period\n<!-- vale Google.Exclamation = NO -->\n<!-- vale Microsoft.Negative = NO -->\n\nRead the [catalog guide](../AGENTS.md) first. It lists the quirks that all\nthree files share, the hazard thresholds, and a recipe that combines the\nreturn periods. The facts below come from this collection's COG. The HTML\ncomment beside each fact identifies the CI check that tests it.\n<!-- guidance -->\n\n## The file\n\nThe `data` asset is `csiro_wf_max_fwi_rp10.tif`, at\n`https://data.sourc","encoding":"utf8","key":"portolan-mirrors/wb-fwi/fwi-rp10/AGENTS.md","truncatedAt":512},{"bytes":"# Wildfire hazard: FWI, 10-year return period\n\n<!-- vale Google.Exclamation = NO -->\n\nThe yearly maximum Fire Weather Index (FWI) at noon local time, expected once in 10 years,\non a global 0.5° grid. Higher values mean weather that favors severe fire.\n\n![FWI, 10-year return period](thumbnail.png)\n\n## Method\n\n[CSIRO](https://www.csiro.au/) fit a GEV distribution to the yearly maxima of\ndaily FWI in NASA's\n[Global Fire Weather Database](https://data.giss.nasa.gov/impacts/gfwed/), from\n1981 to 2016. This grid","encoding":"utf8","key":"portolan-mirrors/wb-fwi/fwi-rp10/README.md","truncatedAt":512}],"sampleObjects":["portolan-mirrors/wb-fwi/fwi-rp10/AGENTS.md","portolan-mirrors/wb-fwi/fwi-rp10/README.md","portolan-mirrors/wb-fwi/fwi-rp10/collection.json","portolan-mirrors/wb-fwi/fwi-rp10/csiro_wf_max_fwi_rp10.tif","portolan-mirrors/wb-fwi/fwi-rp10/thumbnail.png"],"sidecars":[{"bytes":"# World Bank global wildfire hazard\n\n<!-- vale Google.Exclamation = NO -->\n\nA Portolan mirror of the World Bank's\n[global wildfire hazard](https://datacatalog.worldbank.org/search/dataset/0066912/global-wildfire-hazard)\ndataset, published as cloud-optimized GeoTIFFs.\n\n## What is here\n\nThe catalog has one collection per return period. Each collection stores the yearly maximum Fire Weather\nIndex (FWI) expected once in that period, on a global 0.5° grid.\n[CSIRO](https://www.csiro.au/) computed the grids for the\n[ThinkHazard!](https://thinkhazard.org/) wildfire layer, from daily FWI in\nNASA's [Global Fire Weather Database](https://data.giss.nasa.gov/impacts/gfwed/)\nfor 1981 to 2016.\n\n- [FWI, 2-year return period](fwi-rp2/README.md)\n- [FWI, 10-year return period](fwi-rp10/README.md)\n- [FWI, 30-year return period](fwi-rp30/README.md)\n\n## License\n\n[CC BY 4.0](https://creativecommons.org/licenses/by/4.0/), as the World Bank\nData Catalog lists it. Attribute the World Bank and CSIRO when you reuse the\ndata.\n\n## Provenance\n\nThis catalog is a mirror, not the official publication. The World Bank\npublishes the original GeoTIFFs and the technical basis PDF. The World Bank\nlists the dataset as \"No further updates planned\". The mirror converts each\nGeoTIFF to a COG with embedded statistics and leaves every pixel value\nunchanged. The build is in\n[nlebovits/wb-fwi](https://github.com/nlebovits/wb-fwi). It pins each source\nfile by SHA-256, so a changed upstream file stops the build.\n\n## Access\n\nRead the 2-year return level at one point with\n[rasterio](https://rasterio.readthedocs.io/):\n\n```python\nimport rasterio\n\nURL = (\"https://data.source.coop/portolan-mirrors/wb-fwi\"\n       \"/fwi-rp2/csiro_wf_max_fwi_rp2.tif\")\nwith rasterio.open(URL) as src:\n    print(next(src.sample([(150.3, -33.7)]))[0])\n```\n\nRead the [agent guide](AGENTS.md) before you query. It lists the quirks that\nproduce wrong answers. One example: 0 marks masked oceans and deserts as well\nas cells with no fire weather.\n","encoding":"utf8","key":"portolan-mirrors/wb-fwi/README.md","name":"README.md","size":1996,"truncated":false},{"bytes":"{\n  \"type\": \"Catalog\",\n  \"stac_version\": \"1.1.0\",\n  \"stac_extensions\": [\n    \"https://schemas.portolan-sdi.org/portolan/v0.2.0/schema.json\"\n  ],\n  \"id\": \"wb-fwi\",\n  \"title\": \"World Bank global wildfire hazard\",\n  \"description\": \"A mirror of the World Bank's global wildfire hazard dataset. CSIRO computed it for [ThinkHazard!](https://thinkhazard.org/) from NASA's Global Fire Weather Database. Three collections hold Fire Weather Index return levels for 2, 10, and 30-year return periods on a global 0.5° grid, as cloud-optimized GeoTIFFs. Read [AGENTS.md](AGENTS.md) before you query.\",\n  \"links\": [\n    {\n      \"rel\": \"root\",\n      \"href\": \"./catalog.json\",\n      \"type\": \"application/json\",\n      \"title\": \"World Bank global wildfire hazard\"\n    },\n    {\n      \"rel\": \"describedby\",\n      \"href\": \"./README.md\",\n      \"type\": \"text/markdown\",\n      \"title\": \"Catalog README\"\n    },\n    {\n      \"rel\": \"agents\",\n      \"href\": \"./AGENTS.md\",\n      \"type\": \"text/markdown\",\n      \"title\": \"Catalog agent guide\"\n    },\n    {\n      \"rel\": \"self\",\n      \"href\": \"https://data.source.coop/portolan-mirrors/wb-fwi/catalog.json\",\n      \"type\": \"application/json\",\n      \"title\": \"World Bank global wildfire hazard\"\n    },\n    {\n      \"rel\": \"child\",\n      \"href\": \"./fwi-rp2/collection.json\",\n      \"type\": \"application/json\",\n      \"title\": \"Wildfire hazard: FWI, 2-year return period\"\n    },\n    {\n      \"rel\": \"child\",\n      \"href\": \"./fwi-rp10/collection.json\",\n      \"type\": \"application/json\",\n      \"title\": \"Wildfire hazard: FWI, 10-year return period\"\n    },\n    {\n      \"rel\": \"child\",\n      \"href\": \"./fwi-rp30/collection.json\",\n      \"type\": \"application/json\",\n      \"title\": \"Wildfire hazard: FWI, 30-year return period\"\n    },\n    {\n      \"rel\": \"vcs\",\n      \"href\": \"https://github.com/nlebovits/wb-fwi\",\n      \"title\": \"Source repository\"\n    },\n    {\n      \"rel\": \"issues\",\n      \"href\": \"https://github.com/nlebovits/wb-fwi/issues\",\n      \"title\": \"Issue tracker\"\n    }\n  ],\n  \"updated\": \"2026-10-12T00:00:00Z\"\n}\n","encoding":"utf8","key":"portolan-mirrors/wb-fwi/catalog.json","name":"catalog.json","size":2028,"truncated":false},{"bytes":"# Wildfire hazard: FWI, 10-year return period\n\n<!-- vale Google.Exclamation = NO -->\n\nThe yearly maximum Fire Weather Index (FWI) at noon local time, expected once in 10 years,\non a global 0.5° grid. Higher values mean weather that favors severe fire.\n\n![FWI, 10-year return period](thumbnail.png)\n\n## Method\n\n[CSIRO](https://www.csiro.au/) fit a GEV distribution to the yearly maxima of\ndaily FWI in NASA's\n[Global Fire Weather Database](https://data.giss.nasa.gov/impacts/gfwed/), from\n1981 to 2016. This grid gives the FWI value that the fit expects once in 10\nyears. [ThinkHazard!](https://thinkhazard.org/) combines this grid with the\n[2-year](../fwi-rp2/README.md) and\n[30-year](../fwi-rp30/README.md) grids into four wildfire hazard\nclasses. The [technical basis](https://datacatalogfiles.worldbank.org/ddh-published/0066912/DR0095597/THhazardclass_Wildfire_CSIRO.pdf) gives the method and the thresholds.\n\nGFWED masks oceans, large lakes, and deserts. Those cells hold 0. The\n[agent guide](AGENTS.md) and the [catalog guide](../AGENTS.md) list this quirk\nand the others, with recipes that run.\n\n## Files\n\n| Asset | What it is |\n|---|---|\n| `data` | `csiro_wf_max_fwi_rp10.tif`, a Float32 COG with embedded statistics |\n| `source` | The World Bank original GeoTIFF, resource DR0095655 |\n| `technical_basis` | The CSIRO technical basis PDF |\n| `thumbnail` | The preview above |\n\n## License\n\n[CC BY 4.0](https://creativecommons.org/licenses/by/4.0/), as the\n[World Bank Data Catalog](https://datacatalog.worldbank.org/search/dataset/0066912/global-wildfire-hazard) lists it. Attribute the World Bank and CSIRO.\n\n## Provenance\n\n- **Producer:** CSIRO, for the World Bank's ThinkHazard! tool, in 2017.\n- **Publisher:** the World Bank, [dataset 0066912](https://datacatalog.worldbank.org/search/dataset/0066912/global-wildfire-hazard). The World Bank\n  lists the dataset as \"No further updates planned\".\n- **Mirror:** [portolan-mirrors/wb-fwi](https://source.coop/portolan-mirrors/wb-fwi)\n  on Source Cooperative, built by [nlebovits/wb-fwi](https://github.com/nlebovits/wb-fwi).\n  The build converts the GeoTIFF to a COG with DEFLATE compression and\n  embedded statistics. It changes no pixel value.\n","encoding":"utf8","key":"portolan-mirrors/wb-fwi/fwi-rp10/README.md","name":"README.md","size":2203,"truncated":false}],"sourceObjectKey":"portolan-mirrors/wb-fwi/fwi-rp10/AGENTS.md","starterCell":"with open('/mnt/archive/fwi-rp10/AGENTS.md', 'rb') as fh:\n    head = fh.read(1024)\nprint(len(head), head[:64])","starterSamplePath":"fwi-rp10/csiro_wf_max_fwi_rp10.tif","totalObjects":"3 top-level prefixes"},"summary":"Global wildfire hazard grids from the World Bank, computed by CSIRO for the ThinkHazard! wildfire layer: for each 0.5 degree cell, the yearly maximum Canadian Fire Weather Index…","title":"World Bank Global Wildfire Hazard: Fire Weather Index (FWI) Return Levels, 2, 10 and 30 year, 0.5 degree (COG)","updatedAt":"2026-11-18T19:03:42.391Z","visibility":"public","websiteUrl":"https://datacatalog.worldbank.org/search/dataset/0066912/global-wildfire-hazard"}}