Criora climate risk

Everything about one to ten places in one call: each day's risk on the dates asked (the 7-day forecast, then ECMWF ensemble chances as far as its run reaches, about two weeks, then the month's normal), hazards happening nearby, the headline long-term exposures, a home's damage band and the country's risk. headline is the line that matters most. Use it first for open questions: a trip ("I'm flying to Tokyo on the 25th": one call with every stop and its dates), moving or buying a home ("is it safe to live in..."), "what are the risks in...". For one specific question the matching tool answers in full.

Two to ten named countries side by side: overall risk score, class and world rank, and their climate, nature, social, governance and adaptation scores (0 to 100, higher is riskier; rank 1 is the riskiest of 194). To find the riskiest countries use rank_countries.

Indicative damage to a building or asset at a place from river and coastal flooding and tropical cyclone wind: expected annual damage as a share of its value and in money, and the 50 to 500-year events, low, central and high; today, or RCP4.5 or RCP8.5 in 2050 and 2080 (wind to 2050). Use for what flooding could cost a home, a property or loan collateral, and ESRS E1, IFRS S2, TCFD or EU Taxonomy screening. Give asset_value for money in its currency, which any one building needs; without it the money is for all buildings of its class in the point's 1 km cell, in euros, never one building's. A town or city asked by name resolves to one point of it: `built_ground` then says how its built ground shares out among the bands, which is the answer for the place; the money and band are the point's. Indicative, not a valuation: pass the caveats on.

One Criora document as markdown, with the page to cite, by an id search returned: country:<ISO3>, place:<latitude>,<longitude>, page:methodology, page:sources, catalog:layers or catalog:indicators.

Candidate places for a name, best match first, each with coordinates, full name and kind (city, island, volcano...). Call it only when a name is ambiguous (several Springfields) or one place is used in several calls: every place tool takes a name itself and uses its best match. GeoNames gazetteer in every language; no street addresses.

The next 1 to 7 days' weather danger over a whole country or region, day by day: the worst class reached, its hazard, the share of the area at each class, the worst point and the cities most at risk; with categories or layers, also long-term climate layers over the area (mean, 10th and 90th percentile, share at or above a value). Use for "how bad is the weather in Japan this week", "how much of Spain faces high heat tomorrow", "how much of Bangladesh floods in a 100-year flood". One place: get_forecast or get_climate_profile.

Long-term climate risk of one place or property in depth, from 57 global layers: river and coastal flood depth of a 100-year flood now, 2050 and 2080; sea level rise; extreme heat and cold days; heavy rain; drought; water stress; wildfire; warming and rainfall change to 2050 and 2080 with the models' range; NDVI vegetation; earthquake hazard, subsidence, elevation, land cover and the climate zone. Narrow it with categories or layer ids (fetch catalog:layers). Use for buying, renting or building a home or moving somewhere: "does this area flood", "how hot will it get here by 2050". Two or more places, a list of sites or a portfolio: get_climate_table, up to 200 in one call. Not for this week's weather (get_forecast) or events now (get_hazards_near).

A list of sites, a portfolio, addresses or spreadsheet rows screened in one call: the long-term layers of get_climate_profile at up to 200 places, as a table, a row per place and a column per layer and band. By default the headline exposures: river and coastal flood depth, sea level rise, heat, heavy rain, water stress, drought, wildfire, earthquake hazard and the climate zone; narrow it with categories (flooding...) or layer ids. Each row carries a home's damage band there. Use it whenever two or more places are asked about, rather than a profile per place: rank sites on the bands and columns, or compare places on the same layers. One place in depth: get_climate_profile.

The large city whose climate today is most like a city's climate in 2041-2060, its climate twin: "what will Belgrade feel like in 2050". Under SSP2-4.5 and SSP5-8.5, with how alike the climates are and how much warmer and wetter or drier the city gets. For large cities; seasonal means, not extremes. Long-term exposure of a place: get_climate_profile.

One country's climate and disaster risk, of 194: overall score 0 to 100 (higher is riskier), class and world rank, and its climate, nature, social, governance and adaptation scores. response_format=detailed adds the INFORM, ND-GAIN and World Bank indicators behind them. Use for one country in depth, ESG or investment screening, "how risky is Paraguay". Named countries side by side: compare_countries; the most or least at risk: rank_countries.

El Niño and La Niña now and ahead: NOAA CPC's alert status and its chances of each phase for the nine seasons ahead. With a country, what each phase has brought it in each season since 1950 (its wetter or drier, warmer or colder third): tendencies of the past, not a forecast. Use for "is El Niño coming", "will El Niño make Indonesia dry this winter".

This week's weather and extreme-weather risk at a place, day by day: temperature, rain, wind and each day's risk class (Very low, Low, Medium, High, Critical) with the hazard behind it and why. Heat and cold are graded on UTCI, the felt temperature outdoors. Use for the next 7 days: "will it storm in Lisbon on Friday", "is it too hot to hike this week". Days 8 and later: assess_place with dates. Disasters already happening: get_hazards_near. Long-term exposure: get_climate_profile.

Disasters, hazards and official warnings happening now or recently near a place: wildfires, earthquakes, tsunamis, tropical cyclones with cone and track, floods, eruptions and volcano alert levels, ash clouds, droughts, official weather warnings and air quality. Each has Criora Severity (severity_rank, 0 to 100) beside the issuer's own level, its distance, and for a warning whether the place is inside it. Use for "what is happening in or near X", "what happened in Valencia", "is it safe to go there now". Narrow with radius_km, event_types, min_severity or days (up to 30). Forecast: get_forecast; long-term risk: get_climate_profile.

One hazard event the interactive map opens when the reader taps a warning's area or a storm's cone: the event, or none when it is no longer listed. The map calls it in clients that display it; to list the hazards near a place, use get_hazards_near.

The hazard events the interactive map loads when the reader pans or zooms to a new view. The map calls it in clients that display it; to list the hazards near a place, use get_hazards_near.

The official warnings the interactive map draws over a view: each hazard's edge at each class and its symbols, at the view's zoom. The map calls it in clients that display it; to list the warnings near a place, use get_hazards_near.

What has hit a place before: "has Lisbon had a big earthquake", "how often do typhoons reach Manila", "what floods has Valencia had". Earthquakes, tropical cyclones, tsunamis, floods and eruptions near it: how many by decade and the largest, with date, distance, deaths and damage. A count is complete only from `complete_from` (quakes of M5 and up 1976, cyclones 1980, floods 2021, and its `note` says so: a rise in floods before then is no trend). Ask for more years to reach the great events: the record goes back to 1700 for those that cost lives, and to 1870 for Europe's floods. What is happening now: get_hazards_near.

Disasters, hazards and official warnings across a country, a region (bbox) or the whole world, most severe first by Criora Severity (severity_rank) beside the issuer's own level: wildfires, earthquakes, tsunamis, tropical cyclones, floods, eruptions, volcano alert levels, ash clouds, droughts, weather warnings and air quality. Use for "what is happening in Japan", "which red warnings are in force in Europe", "the strongest earthquakes this week" (days 7), what happened in a country over recent days (days, up to 30). Near one place: get_hazards_near. A country holds an event by its point. For the whole world the answer opens with the Earth's pulse (earth_pulse), one figure for "how dangerous is the world today".

The cities (of 100,000 or more) or the countries where the next 1 to 7 days' weather is most dangerous, ranked by how high the risk reaches and how many people it reaches: heat, cold, wind, heavy rain, snow and freezing rain. Use for "where will it be hottest this week", "which European cities face storms in the next two days", "which countries have the most people at high risk". One place: get_forecast; disasters now: list_hazards; countries' long-term risk: rank_countries.

The countries most or least at climate and disaster risk, overall, on one dimension or on one hazard or other indicator, optionally within a region, subregion or income group, each with its score and world rank (a filter narrows the list, not the rank). Use for "the most climate-exposed countries", "the riskiest countries in Southern Asia", "which countries are most exposed to river flooding", "where do cyclones reach the most people". Named countries: compare_countries; this week's weather: rank_by_forecast.

Find Criora's documents to read and cite with fetch: a country's risk profile, a place's report, the methodology and data sources pages, and the data catalogue (climate layers, NDVI, country indicators). For an answer in one call the specific tools are shorter.

Shows an interactive map in clients that render apps (Claude, ChatGPT): this week's forecast risk, hazard events and official warnings for a place, several places side by side, one event (its area, a storm's cone and track) or the whole world; layer opens it on rain, temperature or wind instead. A client without apps gets a text summary and a link. Use it when the user wants to see a place, a storm or the world on a map.