Partly live.Values marked LIVE come from NOAA, the Bureau of Meteorology, Copernicus (ERA5) and NASA, each with its date and source. Anything marked FIXTURE — including demo events — is still a placeholder.Sources
Climate
Explore how Earth's climate system works, see what is unusual right now, and understand the trends shaping our future — every number with its baseline.
Globe: NASA Blue Marble, Black Marble and MODIS cloud composite — decorative, not today's weather.
Daily analysis · GHRSST MUR · baseline 2003–2014
Sea surface temperature anomaly (°C)
This layer's anomaly is relative to the provider's 2003–2014 MUR climatology — not the WMO 1991–2020 normal used elsewhere on this page.
2 m air temperature · MERRA-2 reanalysis · absolute values
Air temperature, monthly (reanalysis) (K)
Absolute monthly mean, not an anomaly. Reanalysis months are published with a lag.
Still pending: Air-temperature and rainfall anomalies against 1991–2020 need a reanalysis source (candidate: ERA5 via Copernicus C3S). They stay empty until computed from real data.
Monthly global averages and long-term context (1991–2020)
Variable: Global mean surface air temperature · Unit: °C · Period: Last 12 months · Baseline: 1991–2020
Global mean surface temperature
Monthly average, 1991–2020 climatology and recent years
| Time | Current year (2026) | Previous year (2025) | 1991–2020 mean |
|---|---|---|---|
| 2026-01-15 | 13.0 | 13.2 | 12.4 |
| 2026-02-15 | 13.3 | 13.4 | 12.7 |
| 2026-03-15 | 13.9 | 14.1 | 13.4 |
| 2026-04-15 | 14.9 | 15.0 | 14.4 |
| 2026-05-15 | 15.8 | 15.8 | 15.3 |
| 2026-06-15 | 16.5 | 16.5 | 16.0 |
| 2026-07-15 | 16.9 | 16.7 | 16.2 |
| 2026-08-15 | 17.0 | 16.6 | 16.1 |
| 2026-09-15 | 16.3 | 16.1 | 15.5 |
Source: Copernicus C3S / ECMWF · ERA5 · Climate PulseReanalysisUpdated: 26 Sept 2026, 12:00 UTCMethodology
How extreme is it compared with the historical record?
TemperatureLive: Updated from a connected data source.
Above every 1991–2020 year
Global 30-day mean to 26 Sept 2026 vs the same window in 1991–2020 (ERA5)
—
Return period
Precipitation
— percentile
Pending a reanalysis source
—
Return period
Percentile. Ranks a value within the reference period. A 98th-percentile month is warmer than 98% of the same calendar months in 1991–2020. rank ÷ number of years in the baseline × 100
Return period. The average interval between events at least this extreme. A “1-in-50-year” event has about a 2% chance of happening in any given year — it can happen twice in a decade. T ≈ 1 ÷ p (annual exceedance probability)
Both depend on the baseline and on the method used to fit the distribution; a warming climate shifts them over time.
Anomalies relative to 1991–2020 · Last 12 months
| Region | Temperature anomaly | Status |
|---|---|---|
| Global | +0.84 °C | Live |
| Asia | — | Unavailable |
| Europe | — | Unavailable |
| Africa | — | Unavailable |
| North America | — | Unavailable |
| South America | — | Unavailable |
| Oceania | — | Unavailable |
| Maritime Continent | — | Unavailable |
Reference period: 1991–2020
Key ocean and atmosphere indicators that influence global and regional climate.
El Niño–Southern Oscillation
+1.80 °C
El Niño conditions · JJA 2026
Coupled ocean–atmosphere variability in the tropical Pacific, commonly tracked with the Oceanic Niño Index (Niño 3.4 SST anomaly).
Oceanic Niño Index: 3-month mean Niño-3.4 SST anomaly. CPC's thresholds are ±0.5 °C; an official El Niño or La Niña episode needs five overlapping seasons past the threshold.
NOAA Climate Prediction Center · Oceanic Niño Index (ERSSTv5)
SST anomaly, tropical Pacific · 28 Sept 2026 · vs 2003–2014
Indian Ocean Dipole
+0.15 °C
Neutral range · May 2026
Baseline not stated by the provider (NOAA PSL) — pending verification.
The difference in sea surface temperature anomaly between the western and south-eastern tropical Indian Ocean (Dipole Mode Index).
Dipole Mode Index from HadISST1.1 via NOAA PSL, a monthly series published with a lag of a few months. BoM uses ±0.4 °C as event thresholds.
NOAA Physical Sciences Laboratory · Dipole Mode Index (HadISST1.1)
SST anomaly, Indian Ocean · 28 Sept 2026 · vs 2003–2014
Madden–Julian Oscillation
Phase 1
Active · amplitude 1.29 · Western Hemisphere and Africa · 22 Sept 2026
An eastward-moving pulse of tropical convection and circulation that recurs every 30–60 days, tracked with the RMM index (phase and amplitude).
Real-time Multivariate MJO index (RMM) from the Bureau of Meteorology. Amplitude ≥ 1 is conventionally treated as an active MJO; the phase gives where its enhanced convection sits.
Bureau of Meteorology (Australia) · Real-time Multivariate MJO index (RMM)
Daily precipitation, Indo-Pacific · 28 Sept 2026
Atlantic Multidecadal and Pacific Decadal indices will follow once their source series are connected.
Long-term changes in Earth's climate system.
Annual average, relative to 1991–2020
Variable: Annual global mean surface air temperature anomaly · Unit: °C · Period: 1940–2025 · Baseline: 1991–2020
| Time | Annual anomaly |
|---|---|
| 1940-07-01 | -0.69 |
| 1941-07-01 | -0.64 |
| 1942-07-01 | -0.73 |
| 1943-07-01 | -0.73 |
| 1944-07-01 | -0.54 |
| 1945-07-01 | -0.67 |
| 1946-07-01 | -0.67 |
| 1947-07-01 | -0.63 |
| 1948-07-01 | -0.69 |
| 1949-07-01 | -0.65 |
| 1950-07-01 | -0.77 |
| 1951-07-01 | -0.63 |
| 1952-07-01 | -0.65 |
| 1953-07-01 | -0.57 |
| 1954-07-01 | -0.77 |
| 1955-07-01 | -0.79 |
| 1956-07-01 | -0.85 |
| 1957-07-01 | -0.60 |
| 1958-07-01 | -0.53 |
| 1959-07-01 | -0.57 |
| 1960-07-01 | -0.61 |
| 1961-07-01 | -0.54 |
| 1962-07-01 | -0.60 |
| 1963-07-01 | -0.56 |
| 1964-07-01 | -0.77 |
| 1965-07-01 | -0.71 |
| 1966-07-01 | -0.63 |
| 1967-07-01 | -0.62 |
| 1968-07-01 | -0.68 |
| 1969-07-01 | -0.55 |
| 1970-07-01 | -0.58 |
| 1971-07-01 | -0.73 |
| 1972-07-01 | -0.59 |
| 1973-07-01 | -0.49 |
| 1974-07-01 | -0.77 |
| 1975-07-01 | -0.74 |
| 1976-07-01 | -0.81 |
| 1977-07-01 | -0.53 |
| 1978-07-01 | -0.58 |
| 1979-07-01 | -0.43 |
| 1980-07-01 | -0.30 |
| 1981-07-01 | -0.26 |
| 1982-07-01 | -0.46 |
| 1983-07-01 | -0.27 |
| 1984-07-01 | -0.48 |
| 1985-07-01 | -0.52 |
| 1986-07-01 | -0.43 |
| 1987-07-01 | -0.28 |
| 1988-07-01 | -0.25 |
| 1989-07-01 | -0.36 |
| 1990-07-01 | -0.13 |
| 1991-07-01 | -0.19 |
| 1992-07-01 | -0.42 |
| 1993-07-01 | -0.37 |
| 1994-07-01 | -0.34 |
| 1995-07-01 | -0.16 |
| 1996-07-01 | -0.29 |
| 1997-07-01 | -0.17 |
| 1998-07-01 | 0.02 |
| 1999-07-01 | -0.25 |
| 2000-07-01 | -0.25 |
| 2001-07-01 | -0.09 |
| 2002-07-01 | 0.01 |
| 2003-07-01 | -0.01 |
| 2004-07-01 | -0.07 |
| 2005-07-01 | 0.09 |
| 2006-07-01 | 0.04 |
| 2007-07-01 | 0.04 |
| 2008-07-01 | -0.10 |
| 2009-07-01 | 0.04 |
| 2010-07-01 | 0.13 |
| 2011-07-01 | -0.01 |
| 2012-07-01 | 0.04 |
| 2013-07-01 | 0.07 |
| 2014-07-01 | 0.11 |
| 2015-07-01 | 0.26 |
| 2016-07-01 | 0.44 |
| 2017-07-01 | 0.34 |
| 2018-07-01 | 0.26 |
| 2019-07-01 | 0.40 |
| 2020-07-01 | 0.43 |
| 2021-07-01 | 0.28 |
| 2022-07-01 | 0.30 |
| 2023-07-01 | 0.60 |
| 2024-07-01 | 0.72 |
| 2025-07-01 | 0.59 |
Source: Copernicus C3S / ECMWF · ERA5 · Climate PulseReanalysisUpdated: 26 Sept 2026, 12:00 UTCMethodology
Global
—
Trend not calculated
Asia
—
Trend not calculated
Europe
—
Trend not calculated
Africa
—
Trend not calculated
North America
—
Trend not calculated
South America
—
Trend not calculated
A trend is only meaningful with its period stated. When connected, trends are reported as °C per decade by least-squares fit over a named period, with uncertainty. No trend is calculated from fixture data.
Put current conditions into perspective with historical data, comparisons and real-world impacts.
Over the 30 days to 26 Sept 2026, the global mean near-surface air temperature was +0.84 °C relative to 1991–2020 — warmer than 100% of the same 30-day window in 1991–2020 (ERA5). One month is weather-scale variability; the long-term trend is in the chart above.
Read the methodCompared to the 20th century
—
Pending data
Compared to last year
—
Pending data
Similar past events
—
Pending data
01
Take the observed or analysed value for a place and time.
02
Set it against the same calendar period in the 1991–2020 climatology.
03
Express it as an anomaly and a percentile of the baseline distribution.
04
Where the data allow, estimate a return period with its uncertainty.
05
Connect it to the drivers — circulation, climate modes, long-term warming.
One unusual month is weather variability. A shift in the whole distribution over decades is a climate trend. The two are kept separate here.
Climate data for a more resilient world
From historical records to near-real-time monitoring — planned for developers, researchers and decision-makers.
