ClimateData story1955–2025

The storm count is the wrong metric.

Climate change has not produced a clear rise in the number of tropical cyclones. It is loading the storms that do form with more rain, greater odds of major intensity, less warning time and a higher sea.

By Kalaivani Chandramohan · Published August 12, 2026

86.7

tropical cyclones a yearAverage annual count in the complete 1981–2024 IBTrACS window, using storms that reached at least 34 knots in the harmonized U.S. wind field.

01 — The count

The annual total moves sideways, not steadily up.

Five-year averages fluctuate with El Niño, La Niña and ocean-basin cycles. But the first block averaged 85 storms; the latest complete block averaged 88. Counting alone misses the change that matters.

Global tropical cyclones by five-year period

Mean annual count; whisker shows annual low–high

Storms are unique IBTrACS main tracks whose maximum USA_WIND reached at least 34 knots. The incomplete 2025 season in the downloaded archive was excluded. Wind averaging and agency practices still vary, so this is a climatological orientation rather than a homogeneous intensity record. Source: NOAA/NCEI IBTrACS v4r01.

02 — The intensity

A stable count can hide a more dangerous distribution.

The strongest observational result is conditional: given that a cyclone forms, the odds that it reaches major intensity have risen. The satellite era also shows more rapid-intensification events, while confidence in a long-run trend for total cyclone frequency remains low.

Major-cyclone probability

+8% / decade

Estimated increase from 1979–2017 in a homogenized satellite analysis; 95% confidence interval: +2% to +15% per decade.

Rapid intensification

↑

The global frequency of rapid-intensification events likely increased over the past four decades.

IPCC: likely

All-category frequency

?

There is low confidence in a long-term global trend. Models generally project little change or a decrease in total formation.

IPCC: low confidence

The +8% estimate is an increase in the probability that a cyclone exceeds the major-storm threshold, not an 8-percentage-point rise or an increase in the number of storms. Sources: Kossin et al. (2020), NOAA repository; IPCC AR6 WGI Chapter 11.

03 — The rain

The tail of the rain distribution moves faster than the mean.

Global-average precipitation is constrained by Earth’s energy balance. Extreme rainfall is constrained more directly by moisture: a warmer atmosphere can hold about 7% more water vapour per degree Celsius.

Assessed precipitation response per 1°C of warming

Approximate global rate; not a forecast for every location

  • Global mean precipitationVery likely range: 1–3% ~2%
  • Heavy precipitation intensityTracks atmospheric moisture capacity ~7%
  • Peak tropical-cyclone rain ratesCan exceed the moisture-only response ≥7%
Greenhouse gases trap heatHuman influence is the main driver of observed warming.
Air and ocean warmMore evaporation, more atmospheric moisture and a larger ocean heat reservoir.
Storms can release more waterCirculation and wind shear still determine whether, where and how a storm develops.
The forced signal

Greenhouse-gas warming increases atmospheric moisture and ocean heat, making heavier rainfall physically expected.

The noisy geography

ENSO, aerosols and multi-decadal circulation redistribute cyclone activity among basins and years.

The rates compare different physical quantities. They should not be added together or applied mechanically to an individual storm. Observed heavy precipitation has intensified over most land regions with good coverage, and human influence is likely the main driver. Source: IPCC AR6 WGI Chapter 8 and Chapter 11.

04 — The fuel

The upper ocean has accumulated a much larger heat anomaly.

Tropical cyclones do not run on global averages alone. Local sea-surface temperature, subsurface heat, atmospheric stability and wind shear all matter. But the background heat reservoir has shifted unmistakably.

World-ocean heat-content anomaly, upper 700 metres

Five-year means · 1022 joules

Five-year averages of NOAA’s published global 0–700 m anomaly series. The mean for 2016–2025 was 20.95 × 1022 joules above the mean for 1955–1964. An anomaly is a departure from the dataset’s reference climatology, not the ocean’s total heat content. Source: NOAA/NCEI Global Ocean Heat and Salt Content.

05 — The coast

Every coastal storm now begins from a higher baseline.

Rising seas do not create a cyclone. They allow surge and waves to start higher, increasing the probability of compound flooding when coastal water, extreme rain and river flow arrive together.

10 cm

Approximate rise in global mean sea level from 1993 through 2024 in NASA’s satellite record.

2.1 mm/yrEstimated rate in 1993
4.5 mm/yrEstimated rate in 2023
6×

Higher mortality where warning coverage is limited

WMO estimates the disaster-related mortality ratio is nearly six times higher with limited-to-moderate warning coverage than with substantial-to-comprehensive coverage.

30%

Less potential damage with one day of warning

WMO cites an estimate that 24 hours of warning before a storm or heatwave can reduce potential damage by about 30%.

The climate signal changes the hazard. Preparedness changes the outcome.

Sea-level estimates are global means; local relative sea level can differ because of land motion, currents and regional ocean dynamics. Preparedness statistics cover weather-, climate- and water-related hazards rather than tropical cyclones alone. Sources: NASA Sea Level Change Team; WMO early-warning assessment.