July 2026
Hi! Meteorological summer for the Northern Hemisphere is nearing the end, so it’s time for another ‘climate viz of the month’ blog. One of the biggest climate stories right now is El Niño, and I thought it might be a good time to include a visualization that provides another graphical perspective on it while it continues to develop in the central Pacific. Though, to be clear, I expect El Niño to be one of the top climate stories for at least the next 6-12 months.
To better visualize how anomalous the El Niño fingerprint is relative to the background mean warming, I’ve included a graphic this month that removes the global temperature anomaly from a spatial map of surface temperature anomalies. The data I am using are from the 1° gridded product by Berkeley Earth, which incorporates input data sources such as from weather stations, buoys, and ships. I also use their standard reference period of 1951-1980 for computing the anomalies. For those not familiar, the word “anomaly” is a bit of jargon in climate science that simply means a departure from average for a given climate variable.
As anthropogenic climate change continues to warm our planet, temperature anomaly maps look more and more washed out in a sea of red shading. However, temperature gradients are often what give rise to important teleconnections in the climate system, as well as help identify other regional extremes that better reflect actual impacts on society. After all, we don’t live or experience the true global mean. A very simple way to better see these regional differences is by removing the global mean anomaly from each point on the map. Using this method, the red and blue color shading instead shows the relative anomalies compared to the global average. With continued projected warming, I think it might become increasingly helpful for atmospheric scientists to visualize global temperature anomalies by removing the global mean, especially for fields like sea surface temperature, where these spatial patterns and associated gradients are crucial for identifying forecasts of opportunity for subseasonal-to-seasonal (S2S) climate prediction. Note that this concept is similar to the newly adopted approach for measuring the El Niño-Southern Oscillation (ENSO) through the Relative Oceanic Niño Index (RONI), which removes the background mean temperature of the tropics to better isolate the ENSO-related pattern from the long-term tropical warming signal. However, the two approaches are not exactly the same, since my maps remove the global mean anomaly while RONI (using the Niño 3.4 region) removes the tropical mean. Still, both are useful ways of separating large-scale background warming from regional patterns of climate variability, which can be particularly helpful for understanding ENSO teleconnections in the atmosphere (i.e., far-reaching regional climate impacts). Let me know what you think about this approach for visualizing regional climate anomalies! Though, to be clear, I fully recognize that these maps are technical and certainly not the most accessible way of communicating to broad audiences.
Now let’s take a look at July’s pattern of surface temperature anomalies with the global mean removed for 2026 relative to the strong El Niño year of 1997 (July as well). Several features are now clearer in 2026, including the remarkable warmth over land areas of the Northern Hemisphere. This includes the extreme heat in western Europe, but also widespread warm anomalies over North America, Asia, and Africa. Both July 1997 and July 2026 clearly show the imprint of El Niño, with the largest departures in the eastern Pacific in both cases. However, 1997 looks more prominent in this view than 2026. This is in contrast to the current RONI, where both years are at very similar values. The story here is that the 2026 El Niño is occurring in a world that is much warmer due to human-caused climate change. Ocean temperatures overall are also unusually warm, so the El Niño warming is less distinctive from the broader background warmth after the global mean anomaly is removed. This is especially the case since the long-term warming trend of the tropics is closer to the global mean, while many other regions are warming faster, particularly over land areas. There are also more widespread cold departures in July 1997, such as over the Ural Mountains in Siberia, that contributed to a lower global temperature anomaly. Climate change is thus raising the background temperature baseline from which natural variability now occurs, which will most certainly have implications for regional climate impacts compared to other events.
Comparing July 1997 and July 2026, there are also some notable similarities, one of which is the colder departures across Antarctica, as well as the warm signals over Africa and central South America. Some of the biggest differences are across the western United States, western Europe, and western Siberia. Finally, what I think this visualization really demonstrates is just how large of a feature El Niño is, even against the background warmth of global climate change. In short, El Niño is a massive redistribution of heat that sets off teleconnections all around the world while also temporarily boosting global temperatures. Despite this remarkable natural climate phenomenon, we have a lot to learn about El Niño and its effects, and I am 100% confident that this event will bring surprises along the way. Stay tuned. Now let’s turn to the Arctic…
It’s hard to believe that another melt season in the Arctic is already coming to a close. In just a short few weeks from now, Arctic sea ice will reach its annual minimum extent. While there is a clear long-term downward trend due primarily to the emission of greenhouse gases associated with anthropogenic climate change, internal climate variability also plays an important role. In other words, natural variability can temporarily accelerate or slow down shorter-term trends. This is especially evident over the last 15 years or so, where there has not been a clear trend in September Arctic sea-ice extent. Yet again, 2026 will continue the temporary slowdown in late-summer ice melt, with no new records expected. This is primarily due to a very sharp dipole in sea-ice anomalies across the Arctic, with relatively extensive ice cover across the Pacific side of the Arctic and near-record-low conditions across the Atlantic side.
In fact, last month, sea-ice cover in the Beaufort Sea was one of the most extensive years since the 1980s! Eventually, in August, sea ice did start sharply declining in this region, but the focus of this summary is mostly on July. Sea-ice extent was also very high in the Chukchi Sea, making it one of the most ice-filled years again since the 1980s. Much of the Arctic Ocean has been dominated by low pressure over the past few months, which has contributed to ice cover lingering in this region. This was also associated with an arm of thicker multi-year sea ice extending from north of the Canadian Arctic Archipelago to north of Alaska. Since 2013, there have been quite a lot of summers associated with cloudier and stormier conditions, which generally limits surface melt due to less solar radiation reaching the surface. Interestingly, this dipole in sea-ice anomalies this year is also fairly similar to last year.
Changes in mean surface air temperature anomalies (GISTEMPv4; 1951-1980 baseline), mean Arctic sea ice extent (NSIDC; Sea Ice Index v4), and mean Arctic sea ice volume (PIOMAS v2.1; Zhang and Rothrock, 2003) over the satellite era. Updated 8/13/2026.
Despite this extensive sea-ice cover in the Beaufort and Chukchi Seas region, conditions in the Barents and Greenland Seas were at record-low levels in July (see my previous blog on the earliest Barents Sea melt-out). Ice cover in Hudson Bay also had one of the earliest melt-outs on record, with near-surface air temperatures also warmer than average in the region. Looking at the pattern of sea-ice concentration anomalies, areas of open water reached quite far north, including well above 80°N latitude, north of Svalbard. This is even more striking when taking a look at late-August sea-ice concentration, but I’ll talk more about that in next month’s blog.
Near-surface air temperature anomalies reflected this general sea-ice dipole in July, with warmer-than-average conditions toward western Siberia that were more than 5°C above the 1981-2010 average around the Gulf of Ob, and cooler-than-average conditions closer to the North Slope of Alaska. As I discussed last month, temperatures have been cooler closer to the North Pole, associated with the persistent lower pressure (reflected in my 80°N temperature graphics), but relatively warmer to the south and especially over land areas. I’ve seen a lot of climate misinformers sharing that the Arctic had one of the coldest summers on record this year, which appears to come from misinterpreting the 80°N+ graphics as representative of the entire Arctic. But when actually looking at the data for the Arctic Circle (67°N+), it was the warmest June on record and 4th warmest July. Lastly, I sadly still have no news about sea-ice thickness/volume estimates from PIOMAS, so we are flying a bit blind there using that product. Also, if anyone reading has any information on this, I haven’t been able to find any updated sea-ice age data from the Quicklook Arctic Weekly EASE-Grid Sea Ice Age, Version 1 product for 2026. Tough times for trying to find pan-Arctic, year-round data on the health of the ice cover!
That’s it for now. As always, thanks for reading!
Other Blogs (Monthly):
Other Climate Data Statistics (Monthly):
My Visualizations:
My research related to data visualization:
[2] Witt, J.K., Z.M. Labe, A.C. Warden, and B.A. Clegg (2023). Visualizing uncertainty in hurricane forecasts with animated risk trajectories. Weather, Climate, and Society, DOI:10.1175/WCAS-D-21-0173.1
[HTML][BibTeX][Code]
[Blog][Plain Language Summary][CNN]
[1] Witt, J.K., Z.M. Labe, and B.A. Clegg (2022). Comparisons of perceptions of risk for visualizations using animated risk trajectories versus cones of uncertainty. Proceedings of the Human Factors and Ergonomics Society Annual Meeting, DOI:10.1177/1071181322661308
[HTML][BibTeX][Code]
[Plain Language Summary][CNN]
The views presented here only reflect my own. These figures may be freely distributed (with credit). Information about the data can be found on my references page and methods page.