Method and sources
Where every number comes from
This page lists the data behind Cyclones, explained, says which pictures are real data and which are sketches, and explains how the live list works. Figures for Cyclone Arnab are provisional: the agencies revise their tracks and intensities after a storm, and death and damage counts change for days. We will update the page when they do. Spotted a mistake? Tell us through corrections.
Real data, and sketches
| What you see | What it is | Source |
|---|---|---|
| The clouds on the globe in the hero and the 72-hour replay | Real satellite images: infrared at 10.8 µm, one frame an hour on 27 October and every three hours otherwise, 24–28 October 2026. Bright means cold, high cloud tops. Warm sea and land are made transparent so the map shows through. | EUMETSAT, Meteosat Indian Ocean Data Coverage, via the EUMETView map service. Hourly imagery is free to reuse under EUMETSAT's data policy. © EUMETSAT 2026. |
| Arnab's track and the JTWC wind | The Joint Typhoon Warning Center's six-hourly positions and one-minute sustained winds from its warnings, and its final forecast track (dashed). | JTWC warnings for Tropical Cyclone 02B, 26–28 October 2026, as also carried by Weather Underground and GDACS. |
| Arnab's IMD stages and wind | The stage IMD declared and its three-minute wind, read from its bulletins as reported. Times are approximate to within a few hours. IMD's best track, which will replace these, is not yet published. | India Meteorological Department (RSMC New Delhi) bulletins, 24–28 October 2026. |
| The forecast line on the replay chart | IMD's 26 October forecast: a very severe cyclonic storm crossing the coast at 115–125 km/h. The line joins the stages it named; it is not a model run. | IMD national bulletin, 26 October 2026, as reported by Weathernews and ANI. |
| Arnab in numbers | Peak winds from IMD and JTWC; rain from IMD's station report; surge from IMD's forecast; people moved and deaths as reported on 28 October. | IMD; JTWC; Reuters (evacuations); the Andhra Pradesh government, as reported by ThePrint (deaths). |
| Every storm since 1980, and where they were born | All 4,093 storms in IBTrACS from 1980 to 2025, including the 2025 season's provisional tracks, whose winds reached 34 knots (63 km/h), at six-hour steps, coloured by the US agencies' one-minute wind where there is one and the regional agency's otherwise. A birthplace is the first point at or above 34 knots. | NOAA NCEI, IBTrACS v04r01 (Knapp et al., 2010). |
| Sea surface temperature and the 26.5°C line | The 1991–2020 average for each month, on a one-degree grid. | NOAA OISST v2, long-term monthly mean, NOAA PSL. |
| The 3D storm in chapters 03–05 | A sketch, not a weather model. It is built from a cyclone's textbook structure: wind that rises to a peak at the eyewall and falls off outside it (a Holland-type profile), an eyewall that slopes outward with height, spiral bands, an outflow shield, and air that flows in, up and out. Heights are stretched about ten times, and about 18 times when it is cut open. The wind and pressure readout follows the sketch, using the Atkinson–Holliday wind–pressure relation. It is there to show how the parts fit together, not to predict anything. | Structure from Emanuel (2003), Holland (1980) and standard texts. |
| Build a storm | A rule of thumb, not a model. The ceiling is DeMaria and Kaplan's (1994) empirical limit on a storm's one-minute wind from sea temperature: V = 28.2 + 55.8 e0.1813 (SST − 30) m/s. Four factors between 0 and 1 then scale it down: sea temperature (rising from 25.8°C to 27.2°C), wind shear (full above 3 m/s, falling to nothing by 18 m/s), mid-level humidity (from 25% to 80%) and distance from the equator (from 2.5° to 8°). The presets are approximate conditions, chosen so that Arnab lands near its observed peak and Patricia near hers. | DeMaria & Kaplan (1994); Gray (1968). |
| Storm surge | A textbook estimate. The water at the shore is the pressure bulge (about 1 cm per hPa below 1,010 hPa, from the Atkinson–Holliday wind–pressure relation) plus steady wind setup, dη/dx = τ / (ρg(h + η)), with wind stress τ = ρair CD U² (CD = 0.0024), integrated over the real sea-floor profile from 120 km out to the shoreline. Water is drawn inland until the ground rises above it. It leaves out waves, tides, the bend of the coastline and bottom friction. At 45 m/s it gives about 1.8 m at Kakinada, inside IMD's 1.5–2.5 m forecast, and 7–10 m for the Meghna estuary, in line with reports from 1970. | Sea floor: NOAA ETOPO 2022, 60 arc-second, along three straight lines. |
| Overnight | Each storm's six-hourly one-minute wind, with time counted from its first point at 34 kt. "In a day" is the largest gain between two points 24 hours apart. | IBTrACS (NHC best tracks for Otis and Patricia); JTWC warnings for Arnab. |
| Track errors and the cone | IMD's average October–December track forecast errors, 2012–16 and 2017–21. The cone on Arnab's final forecast widens with IMD's long-period average errors (about 72 km at 24 hours). | IMD verification reports. |
| The toll | Official or widely cited deaths: Bhola 1970, 300,000–500,000; Andhra Pradesh 1977, about 10,000–14,200; Bangladesh 1991, 138,866; Odisha 1999, 9,887; Sidr 2007, about 3,400 officially, with estimates up to 10,000; Nargis 2008, 138,373 dead and missing; Phailin 2013, 45; Fani 2019, 64 in Odisha; Amphan 2020, about 128; Arnab 2026, at least 5 so far. | WHO Bulletin (Bangladesh); Government of Odisha; NIDM; Government of Myanmar; reported counts for recent storms. |
| Two seas | Storms reaching 34 kt on JTWC's one-minute wind, counted once each by the sea they formed in, 1982–2000 against 2001–2019. | IBTrACS; Deshpande et al. (2021) for IMD's records. |
| Your coast | Every IBTrACS storm since 1980 with any six-hourly position within 200 km of the place, plus Arnab from JTWC's track. Place search uses Open-Meteo's free geocoding service, from your browser; nothing about you is stored. | IBTrACS; Open-Meteo (CC BY 4.0). |
| The size comparison | Circles showing how far gale-force winds reached: Typhoon Tip about 1,100 km from its centre, Cyclone Tracy about 48 km, Arnab about 130 km on 28 October (JTWC's 34-knot radii). | JTWC; Australian Bureau of Meteorology. |
| The two scales | IMD's seven categories and the Saffir–Simpson scale, with Arnab's reported winds placed on each. | IMD; US National Hurricane Center. |
How the live list works
When you open the page, our server asks two sources which storms are active, at most every 15 minutes:
- NOAA's National Hurricane Center for the Atlantic and the eastern and central Pacific, with each storm's current wind.
- GDACS, the Global Disaster Alert and Coordination System run by the United Nations and the European Commission, for every ocean. GDACS takes its tracks from JTWC and other agencies. Its feed gives each storm's strongest wind so far rather than its current wind, so those storms are shown as "peak so far". A storm stays on the list for 24 hours after its last advisory.
If either source can't be reached, the page says so. The list is for curiosity, not safety. GDACS alerts are generated automatically and are not reviewed by people before they are issued. For warnings, follow your national weather agency.
Words and numbers
- Knots (kt): nautical miles per hour. 1 kt is 1.852 km/h. Kilometre figures are rounded to the nearest 5.
- Sustained wind: the average over one minute (US agencies, including JTWC), three minutes (IMD) or ten minutes (most other agencies). The same storm gives a lower number over a longer average.
- Rapid intensification: an increase of at least 30 kt (about 55 km/h) in 24 hours.
- "Strongest since": we compare peak intensities in IBTrACS using JTWC's winds. On those numbers Arnab (120 kt) is the strongest October–December storm in the Bay of Bengal since Phailin (2013, 140 kt). On IMD's three-minute numbers it is not, because Hudhud (2014) and others rank higher there. Claims that Arnab was the strongest North Indian Ocean storm "since 2019" are not right: Amphan (2020) and Mocha (2023) were both stronger.
Reports and research
- Gray, W. M. (1968). Global view of the origin of tropical disturbances and storms. Monthly Weather Review 96, 669–700.
- Emanuel, K. (2003). Tropical cyclones. Annual Review of Earth and Planetary Sciences 31, 75–104.
- Holland, G. J. (1980). An analytic model of the wind and pressure profiles in hurricanes. Monthly Weather Review 108, 1212–1218.
- Knapp, K. R. et al. (2010). The International Best Track Archive for Climate Stewardship (IBTrACS). Bulletin of the American Meteorological Society 91, 363–376.
- Atkinson, G. D. & Holliday, C. R. (1977). Tropical cyclone minimum sea level pressure / maximum sustained wind relationship for the western North Pacific. Monthly Weather Review 105, 421–427.
- WMO (2020). New list of names for tropical cyclones over the north Indian Ocean.
- JTWC, warning 010 for Tropical Cyclone 02B (Arnab), 28 October 2026.
- Reuters, India evacuates over 180,000 as Cyclone Arnab hits coast, 28 October 2026.
- ThePrint, Cyclone Arnab crosses Andhra coast; damages roads, electric poles, 5 killed in related incidents, 28 October 2026.
- Down To Earth, Cyclone Arnab stalls near Andhra Pradesh coast, 28 October 2026.