iiinova — where your digital self keeps evolving

Tim Osborn @timosbornclim.bsky.social · 24/09/2026

My 3-panel SOI graph inadvertently highlights multidecadal variations in ENSO activity. Relatively active in the top/bottom panels, more quiescent in the middle panel.

1866-present timeseries of Climatic Research Unit SOI with some major El Niño events marked.

161 6

Tim Osborn @timosbornclim.bsky.social · 24/09/2026

The July (Jun-Aug mean) 2026 SOI value exceeds the July values for all the preceding major events: 2026 has the most negative Jun-Aug mean SOI in the 161-year long record. The 2026 El Niño is forecast to strengthen further which could push the SOI to unprecedented year-round values later this year.

Timeseries of 3-month mean SOI (centred on July, so only showing the Jun-Aug mean for each year) with the 2026 value highlighted as being the strongest in the record that starts in 1866.

140 4

Tim Osborn @timosbornclim.bsky.social · 24/09/2026

Aligning by the January closest to peak SOI gives a different view: 2026 El Niño developed in NH summer. 1982-83 did too & then strengthened until its Feb (Jan-Mar) maximum. 1997-98 also developed early in the year, peaked in May (Apr-Jun), then dipped, then reached a higher peak in Mar (Feb-Apr).

SOI overlaid for some major El Niños, each aligned by the January closest to their peak (negative) SOI index.

261 6

Tim Osborn @timosbornclim.bsky.social · 24/09/2026

In the whole timeseries view, 2026 is not (yet) a record breaker. Aligning major events by peak (negative) SOI value, shows 2026 is already very strong though its SOI is not (yet) as strong as it reached in 1982-83, 1877-78, 1905-06, 1997-1998. Final value in 2026 is June-July-August mean.

SOI overlaid for some major El Niños, each aligned by their peak (negative) SOI index.

160 6

Tim Osborn @timosbornclim.bsky.social · 24/09/2026

I’ve updated the long-running (starts 1866) CRU Southern Oscillation Index (SOI). This lets us compare the currently developing El Niño with earlier events including 1877-78. 2026 has developed very quickly, similar to the 1997-98 one. Thick line is 3-month running mean, with y-axis reversed.

24723 47

Tim Osborn @timosbornclim.bsky.social · 23/09/2026

The CRUTEM v5.1 to v5.2 changes are apparent in the 2025 annual mean maps and their difference (USA gap now filled; China cold bias removed; also erroneous NE Canadian station values removed)

Maps of 2025 annual mean temperature anomalies from CRUTEM v5.2 and v5.1 and their difference. Gaps in v5.1 data over the contiguous USA are filled in v5.2. The difference map shows that v5.2 is both more complete and warmer than v5.1 over most of China.

100 1

Tim Osborn @timosbornclim.bsky.social · 23/09/2026

More on the CRUTEM v5.1 to v5.2 changes. 2024-present updated with 380 stations from Chinese Met Admin to remove a cold bias in the data previously used for China. Gap in contiguous USA since spring 2025 that had been triggered by changes to CLIMAT data feed filled using USHCN. Land T difference:

Timeseries from 1850 to 2026 of global average temperature from CRUTEM versions 5.1 and 5.2 and their difference.

110 1

Tim Osborn @timosbornclim.bsky.social · 23/09/2026

New version of the HadCRUT global temperature dataset released @colinmorice.bsky.social HadCRUT5.2 uses HadSST4.2 (with improved adjustments for SST biases during 1933-1961) and CRUTEM5.2 (with updates to stations in China, USA and Canada). Here is the difference from the previous version 5.1

Alt text: Timeseries from 1850 to 2026 of global average temperature from HadCRUT versions 5.1 and 5.2 and their difference.

12410 24

Tim Osborn @timosbornclim.bsky.social · 17/09/2026

And with views over the surrounding landscape

070 7

Tim Osborn @timosbornclim.bsky.social · 17/09/2026

The intriguing unfinished Elizabethan lodge at Lyveden. Apparently it's been standing there, virtually unaltered for >400 years

1111 11

Tim Osborn @timosbornclim.bsky.social · 17/09/2026

There's lots of discussion in public & media about fuel price increases. Useful to know that (at least until the increases of the last week), inflation-adjusted fuel prices in the UK have stayed fairly flat for the last 26 years. Motorists have not been hit harder than the average person.

Timeseries of diesel and unleaded fuel prices in the UK from 2000 to 2026 (year-to-date) for raw prices (which have increased from about 80 pence per litre in 2000 to about 160 pence per litre in August 2026) and for inflation-adjusted prices (which have stayed close to 80 pence per litre in real terms).

154 5

Tim Osborn @timosbornclim.bsky.social · 01/09/2026

Ben Santer et al. show another claim in the Trump Admin's 2025 DOE climate report does not withstand scrutiny. Using the correct observational data, they find that observed tropical atmospheric temperature changes are consistent with the expected human fingerprint on climate. (my annotations)

Fig 3 from Santer et al. (2026) showing consistency between observed and modelled changes in atmospheric temperatures at different levels within the stratosphere and troposphere, with my annotations on panel c

33715 37

Tim Osborn @timosbornclim.bsky.social · 12/08/2026

Who needs pinholes, special glasses or a colander? You just need gaps between the leaves. And a cat. #eclipse

1232 23

Tim Osborn @timosbornclim.bsky.social · 11/08/2026

The CRU scPDSI drought index suggests that global drought area expanded rapidly after mid-2019 and then remained above the long-term mean. Global extent of drought conditions during 2025 declined slightly compared with 2024 and was below the preceding peak observed during the second half of 2023.

Figure show time series of the percentage of global land area (excluding ice sheets and deserts) with self-calibrating Palmer Drought Severity Index (scPDSI) indicating moderate (< –2), severe (< –3), and extreme (< –4) drought for each month of 1950–2025. It shows periods with small drought extents (e.g. late 1950s, around 1980, late 1990s) and periods with large drought extents (e.g. mid 1980s, early 2000s, post 2019).

164 6

Tim Osborn @timosbornclim.bsky.social · 27/07/2026

HadCRUT5 global temperature now updated to May 2026 crudata.uea.ac.uk/cru/data/tem...

Global temperature time series 1850-2026 from HadCRUT5 dataset

1104 10

Tim Osborn @timosbornclim.bsky.social · 26/06/2026

I went sailing on the north Norfolk coast for some relief from the heat. Then the wind changed just after 6pm and the temperature shot up. Here’s Weybourne obs, showing an impressive 8.3C increase in one hour from 6pm to 7pm, reaching 34.8C having not exceeded 29C all day

0110 11

Tim Osborn @timosbornclim.bsky.social · 23/06/2026

At the same time, we've revamped the CRU-TS webpage so it is easier to find the latest version. We've also updated the country average timeseries (called CRU-CY). The Google Earth interface will be updated in the next few weeks, as will the scPDSI drought index. Enjoy!

131 3

Tim Osborn @timosbornclim.bsky.social · 20/06/2026

Found a shady spot on a hot day.

010 1

Tim Osborn @timosbornclim.bsky.social · 12/04/2026

Start Bay: Fantastic setting (+weather) for the micrometeorology day on our first-year undergraduate field course. @ueaenv.bsky.social

171 7

Tim Osborn @timosbornclim.bsky.social · 23/03/2026

Yeah but whatabout the co op. Can't get any more blatant than "free from Easter" 😂

120 2

Tim Osborn @timosbornclim.bsky.social · 11/03/2026

The London Eye from the Artist's Garden

020 2

Tim Osborn @timosbornclim.bsky.social · 25/02/2026

The DoE report was flawed because it only considered the lower stratosphere, and even there it made false claims about model-obs disagreement. Ben's figure shows agreement between model predictions and observations of global temperature change across multiple stratospheric and tropospheric layers.

Figure 1 from Santer et al. (2026) showing agreement between model predictions and observations of global temperature change across multiple stratospheric and tropospheric layers.

120 2

Tim Osborn @timosbornclim.bsky.social · 16/02/2026

You can zoom into a sub period, download the data shown in the time series graph, download the image of the graph. And it is all CC BY, meaning you can use, share and adapt - just acknowledge NCAS (@ncas-uk.bsky.social) and CRU-UEA (@climateuea.bsky.social) as the creators @ueaenv.bsky.social

This Climate Data Dashboard provides timeseries for individual countries and other territories, calculated using version
4.09 of the CRU-TS and CRU-CY datasets.

We remain neutral with regard to any jurisdictional claims.

These datasets have been created by the Climatic Research Unit (CRU) and the National Centre for Atmospheric
Science (NCAS), both at the University of East Anglia (UEA).

You are encouraged to freely use and share these graphs and data if you acknowledge NCAS and CRU-UEA as the creator. If it is appropriate to include a scientific reference, then please also cite "Harris et al. (2020) Scientific Data".

130 3

Tim Osborn @timosbornclim.bsky.social · 16/02/2026

Example: Chile. Seasonal temperatures, March-April-May and June-July-August precipitation and annual drought area

Chile CRU-TS / CRU-CY annual drought area (%)

Chile CRU-TS / CRU-CY seasonal precipitation (mm/month)

Chile CRU-TS / CRU-CY seasonal temperature (degC)

120 2

Tim Osborn @timosbornclim.bsky.social · 16/02/2026

Example: France. Seasonal temperatures, annual precipitation and annual drought area

France CRU-TS / CRU-CY annual drought area (%)

France CRU-TS / CRU-CY annual average precipitation (mm/month)

France CRU-TS / CRU-CY seasonal average temperature (degC)

130 3

Tim Osborn @timosbornclim.bsky.social · 16/02/2026

New dashboard to visualise CRU-TS country averages (i.e. CRU-CY) for temperature, precipitation, humidity and drought area (using our scPDSI indicator). (It takes a little while to first load, but is quite responsive after that) climate.uea.co/CRU-Timeseri...

A screenshot of the CRU-TS national time series dashboard

16229 62

Tim Osborn @timosbornclim.bsky.social · 20/01/2026

One of the many interesting findings in this latest UK Public Attitudes to Science survey

Highlighted quote: “What determines whether someone embraces science? The survey examines the influence of science capital, the sum of one’s interactions with science through work, study, family, friends and leisure, and a guiding principle for the Science Museum Group.

Among those with high science capital, 84% support government funding for blue-skies research and 84% agree science has increased their personal prosperity. Among those with low science capital, just 44% and 13% respectively agree.”

062 6

Tim Osborn @timosbornclim.bsky.social · 14/01/2026

Here's a similar depiction of just HadCRUT5 including its uncertainties: an 1850-1900 baseline but the dataset uncertainties are not re-baselined so that it is clear where the greatest uncertainty lies. The right panel shows the 2025 value plus the uncertainty in its difference from the baseline.

HadCRUT5 global temperature from 1850 to 2025, with 2025 value highlighted. Uncertainty range is included but with offset from 1981-2010 to make it clear where the greatest uncertainty lies.

130 3

Tim Osborn @timosbornclim.bsky.social · 14/01/2026

The right panel of the graph shows 2025 values compared with the 1850-1900 baselines of each dataset individually. This is better for seeing the variation between datasets in the warming from 1850-1900 to 2025, including individual uncertainties where quantified.

2025 values highlighted in a separate panel for each individual dataset, relative to their own individual 1850-1900 baselines (or their own convention for this baseline if they do not cover this period with their own data).

130 3

Tim Osborn @timosbornclim.bsky.social · 14/01/2026

The left-panel of the graph above compares series offset so the average of the 1850-1900 baselines for datasets that begin in 1850 is zero and all series have the same 1981-2010 mean. This allows us to see the greater uncertainty (spread) in the earlier period.

Multi-dataset comparison of global temperatures from 1850-2025. Baseline is a combination of 1850-1900 means.

150 5

Tim Osborn @timosbornclim.bsky.social · 14/01/2026

Most 2025 global temperatures are now out (degrees C above 1850-1900 baseline) 1.41 HadCRUT5 1.44 Berkeley 1.46 JRA-3Q 1.47 Copernicus 1.53 DCENT-I NOAA and NASA GISS values will be public at 2pm UK time and but based on already public Jan-Nov data they will likely be between 1.3 and 1.4 degC.

Multi-dataset comparison of global temperatures from 1850-2025, with 2025 values highlighted in a separate panel for each individual dataset. Baseline is a combination of 1850-1900 means.

69357 93

Tim Osborn @timosbornclim.bsky.social · 21/12/2025

For all you global temperature aficionados, HadCRUT5 is now updated to November. 12-month running mean timeseries and HadCRUT5 Analysis map for November. For the December update -- and release of the 2025 annual temperature data -- put 14 January in your diary.

1185 18

Tim Osborn @timosbornclim.bsky.social · 25/11/2025

I currently have three funded PhD positions available for applications. Links for more info and to apply are in my profile page: research-portal.uea.ac.uk/en/persons/t...

PhD Positions

I have three funded projects currently available.

Closing date 10 December 2025:

Redeveloping Central England Temperature: Modernising a Historic Climate Series (CASE project with Met Office) (main supervisor: Tim Osborn; co-supervisors: Stephen Packman, Met Office, Ed Hawkins, University of Reading, Andrew Friedman, UEA)

Closing date 7 January 2026:

Causes and Syndromes of Extreme Wildfires in Tropical and Extratropical Forests (main supervisor: Matt Jones; co-supervisor: Tim Osborn)

Closing date 7 January 2026:

Predicting and Mitigating Megafires under Climate Change (CASE project with WTW) (main supervisor: Matt Jones; co-supervisor: Tim Osborn)

11312 13

Tim Osborn @timosbornclim.bsky.social · 03/11/2025

HadCRUT5 global temperature dataset now updated to Sep 2025. As usual, data can be downloaded from CRU/UEA website: crudata.uea.ac.uk/cru/data/tem... And from Met Office/hadobs website: www.metoffice.gov.uk/hadobs/hadcr... @climateuea.bsky.social @colinmorice.bsky.social

01710 17

Tim Osborn @timosbornclim.bsky.social · 01/10/2025

Climate stripes by longitude? Here's mine from 2004.

Estimates temperature changes from 1400-1960 over the mid to high latitudes of the Northern Hemisphere as a function of longitude covering North America, Europe and Russia. From work by Tim Osborn and colleagues in 2004, based on records of tree-ring maximum latewood density.

190 9

Tim Osborn @timosbornclim.bsky.social · 22/09/2025

For regular monthly updates to CRUTEM (and hence to HadCRUT) we rely on monthly data provided by National Met Services via the WMO's CLIMAT system. Unusually, the USA data didn't yet make it into the CLIMAT data feed for Mar-Jun 2025. Note the USA-sized gap in the non-infilled version of HadCRUT:

2102 10

Tim Osborn @timosbornclim.bsky.social · 22/09/2025

Thanks! If you zoom in on the PDF version you can see how I drew the different levels of shading 👇

161 6

Tim Osborn @timosbornclim.bsky.social · 22/09/2025

And some visualisations are available here: crudata.uea.ac.uk/~timo/diag/t... including 12-month running means of global land+marine (HadCRUT) and global land (CRUTEM) temperature anomalies

050 5

Tim Osborn @timosbornclim.bsky.social · 22/09/2025

The update to Chinese station data makes a notable difference to the global land temperature data by making data for recent years more consistent with the historical data in the CRUTEM station database (which for China uses the mean of min & max temperature to represent the daily mean temperature)

250 5

Tim Osborn @timosbornclim.bsky.social · 22/09/2025

After a brief (ahem) hiatus, the HadCRUT5 global temperature dataset has now been updated to June 2025 by @metoffice.gov.uk @climateuea.bsky.social and @ncas-uk.bsky.social

1850-2025 time series of global mean temperature anomaly from the HadCRUT5 dataset.

511441 114

Tim Osborn @timosbornclim.bsky.social · 11/09/2025

There’s still time to visit “A Field Hospital for Eco-Anxiety”, open till 28 Sep. It’s an installation at National Trust’s Dunham Massey, put together by “Not here to be liked” including input from UEA / CRU PhD student Erik Hartin. @nationaltrust.org.uk @climateuea.bsky.social @ueaenv.bsky.social

121 2

Tim Osborn @timosbornclim.bsky.social · 11/09/2025

John has a knack for finding the heart of the issue and expressing it in understandable terms.

Screenshot from diagrammonkey blog which says:

"A recent paper on sea level change concludes that “approximately 95% of the suitable locations show no statistically significant acceleration of the rate of sea level rise. The investigation suggests that local, non-climatic phenomena are a plausible cause of the accelerated sea level rise observed at the remaining 5% of the suitable locations.”

Having read the abstract, one can guess – more or less – what the problem is as it’s straight out of the sceptic playbook.

1. Take a global indicator with a clear signal

2. Break it down into as many local estimates as possible, maximising the noise to signal ratio.

3. Test each local change independently using a statistical significance test that is predisposed to accept the null (no change/no accelerations or whatever) and does not account for the possibility of a common signal.

4. Aggregate the results of the individual tests back up to the global scale (again ignoring any possible common signal) and express disbelief that anyone thought there was something there.

It’s a way to avoid seeing the forest by a very careful study of all the individual trees."

161 6

Tim Osborn @timosbornclim.bsky.social · 15/08/2025

UAH *does* agree that 2024 was the hottest year on record. Here's is the excerpt on tropospheric temperature (co-author John Christy from the UAH team). Perhaps your eye is being drawn to the slightly lower temperatures so far in 2025?

Transcript:

5. TROPOSPHERIC TEMPERATURE
—S. Po-Chedley, J. R. Christy, C.-Z. Zou, C. Mears, and L. Haimberger
Record-high global lower tropospheric temperature (LTT) values in the second half of 2023 persisted through boreal summer in 2024 (Fig. 2.9), producing a 2024 global-mean LTT that was 0.84°C (0.77°C–0.92°C depending on dataset) above the 1991–2020 climatological mean (Fig. 2.10). All nine tropospheric temperature datasets considered here (Table 2.3) ranked 2024 as the warmest year for global LTT, making 2023 the second-warmest year in most datasets.

140 4

Tim Osborn @timosbornclim.bsky.social · 15/08/2025

To conclude: climate warming poses a serious threat to coral reefs (as highlighted in IPCC AR6 WG2, text here). Record-breaking one-year declines are alarming and a central part of how the impact plays out, but longer-term records provide essential context and shouldn’t be omitted.

Excerpt from IPCC Sixth Assessment Report, Working Group 2, Chapter 3 Executive summary. Text reads as follows:

Climate impacts on ocean and coastal ecosystems will be
exacerbated by increases in intensity, reoccurrence and
duration of marine heatwaves (high confidence), in some cases,
leading to species extirpation, habitat collapse or surpassing
ecological tipping points (very high confidence). Some habitat forming coastal ecosystems including many coral reefs, kelp forests and
seagrass meadows, will undergo irreversible phase shifts due to marine heatwaves with global warming levels >1.5°C and are at high risk this
century even in <1.5°C scenarios that include periods of temperature
overshoot beyond 1.5°C (high confidence). Under SSP1-2.6, coral
reefs are at risk of widespread decline, loss of structural integrity and
transitioning to net erosion by mid-century due to increasing intensity
and frequency of marine heatwaves (very high confidence). Due to
these impacts, the rate of sea level rise is very likely to exceed that
of reef growth by 2050, absent adaptation. Other coastal ecosystems,
including kelp forests, mangroves and seagrasses, are vulnerable to
phase shifts towards alternate states as marine heatwaves intensify
(high confidence). Loss of kelp forests are expected to be greatest
at the low-latitude warm edge of species’ ranges (high confidence).
{3.4.2.1, 3.4.2.3, 3.4.2.5, 3.4.4}

000 0

Tim Osborn @timosbornclim.bsky.social · 15/08/2025

Southern GBR: 31% one-year decline: “which drove regional coral cover [now at 26.9%] below the long-term average of 29.3%”

1986 to 2025 time series of hard coral cover for the Southern Great Barrier Reef

100 1

Tim Osborn @timosbornclim.bsky.social · 15/08/2025

Central GBR: 14% one-year decline: “however, it [hard cover cover at 28.6%] is still above the long-term average of 19.8%”

1986 to 2025 time series of hard coral cover for the Central Great Barrier Reef

100 1

Tim Osborn @timosbornclim.bsky.social · 15/08/2025

Northern GBR: 25% one-year decline. “Despite this, coral cover [at 30%] remains above the long-term average of 26.6%.”

1986 to 2025 time series of hard coral cover for the Northern Great Barrier Reef

100 1

Tim Osborn @timosbornclim.bsky.social · 14/08/2025

The "Jones" Gibraltar-Iceland NAO index now updated to June 2025. Data and viz here: crudata.uea.ac.uk/cru/data/nao/ @climateuea.bsky.social @ueaenv.bsky.social

091 9

Tim Osborn @timosbornclim.bsky.social · 21/07/2025

Delighted to announce that Ben Santer is now an Honorary Professor @ueaenv.bsky.social & the Climatic Research Unit (CRU). Ben is returning to his "academic home": he did his PhD in CRU before his stellar career developing "fingerprint" methods to identify human & natural effects on climate.

1346 34