All datasets › Weather
Weather in the News 2019–2026: article-level dataset and coverage tables
A dataset of weather news from a global news index, January 2019 – July 2026, built with the Prism News API (historical and real-time news, https://radar-api.fliarbi.com/news). Every row is one news article about the weather: forecasts, warnings, weather services and the effects of hot, cold, wet or severe weather. Each row carries its date, location, tone, editorial priority, themes and the organisations it names, plus a link that returns the full article through the API.
Download
Data-platform copies are being published; links appear here as they go live.
Key findings
In a global news index covering January 2019 to July 2026, January carried the largest share of each year's weather coverage (10.63% on average), and December, January and February together carried 28.35%, against 25% if coverage were spread evenly across the year.
Basis: Within-year month shares of 480,612 matched articles, averaged over the complete years 2019–2025 (within-year shares survive changes in outlet mix).
Weather took a 33% smaller share of news coverage in 2020 than in 2019 (23.14 against 34.71 articles per 10,000 news articles), the lowest annual share between 2019 and 2025, in the year coverage of the COVID-19 pandemic expanded.
Basis: Annual weather articles divided by all articles in the index for the same year; the lowest month was March 2020, at 16.38 per 10,000.
January 2026, the month of the late-January winter storm across North America, was the month in which weather took its largest share of global news between January 2019 and July 2026, at 54.57 articles per 10,000, against 34.18 in December 2025.
Basis: Monthly weather articles per 10,000 news articles, every month from January 2019 to July 2026.
The India Meteorological Department was the organisation named most often in weather news between January 2023 and July 2026, appearing as "IMD" in 32,535 articles (11.17% of all weather articles), ahead of the US National Weather Service (13,243) and the UK Met Office (9,308).
Basis: Organisation names extracted from each article opening, counted once per article; the full name "India Meteorological Department" is a separate label (17,917 articles) that overlaps with the short form.
Of weather articles carrying a tone label between January 2023 and July 2026, 63.9% were negative in tone, 1.9 times the 33.6% negative share across all news in the same index.
Basis: Top sentiment label of each article, grouped into negative, neutral and positive; 99.8% of sampled articles carry a label.
Charts and maps
Coverage by country
Primary location of located articles, January 2023 – July 2026. Lift = the country's share of this topic's news ÷ its share of all located news (above 1 = over-covered).
| Country | Articles | Share | Lift |
|---|---|---|---|
| United States | 60,171 | 39.23% | 0.81 |
| India | 49,267 | 32.12% | 2.43 |
| United Kingdom | 9,282 | 6.05% | 1.04 |
| Australia | 4,400 | 2.87% | 1.34 |
| New Zealand | 2,184 | 1.42% | 2.2 |
| Canada | 1,738 | 1.13% | 0.53 |
| Germany | 1,311 | 0.85% | 0.73 |
| China | 1,306 | 0.85% | 0.32 |
| Japan | 1,231 | 0.8% | 0.76 |
| United Arab Emirates | 1,188 | 0.77% | 1.19 |
| France | 1,106 | 0.72% | 0.37 |
| Spain | 914 | 0.6% | 0.61 |
| Italy | 847 | 0.55% | 0.57 |
| Sri Lanka | 822 | 0.54% | 3.38 |
| Ireland | 794 | 0.52% | 0.72 |
Use the API
This dataset was built with the Prism News API, which serves historical and real-time news. Install the command-line client, log in with an API key, and query the topic yourself:
curl -fsSL https://raw.githubusercontent.com/fliarbi/prism-news-api/main/scripts/install.sh | bash
prism-news login
prism-news trends "imd" --granularity monthly --tableCite
Fliar BI. Weather in the News 2019–2026: article-level dataset and coverage tables. January 2019 – July 2026. https://fliarbi.github.io/news-datasets/weather/