Earth’s Coldest Temperatures: Vostok, Satellite Measurements and Antarctic Hollows

Two very different measurements appear in stories about Earth’s coldest place: air temperature at a station and the temperature of the snow surface estimated from satellite observations.

Station air temperatures and satellite surface measurements

Vostok’s widely recognised record is −89.2°C, measured on July 21, 1983. A 2018 analysis of East Antarctic satellite data identified snow-surface temperatures around −98°C. The second number does not replace the first as an equivalent station air-temperature measurement.

Small hollows can make a difference

The satellite research highlighted extremely dry, clear conditions and shallow depressions where exceptional surface cooling occurred. That adds a useful insight beyond the headline: even an apparently flat polar landscape has topographic variations that matter. These are extreme observations, not typical conditions across all Antarctica.

Further information: NSIDC’s 2018 update on Antarctic surface temperatures

Explore more: cold in inhabited Siberian places; an Antarctic station during blowing snow.

Video: NASA | The Coldest Place in the World (published by NASA Goddard).

Two measurements answer different questions

An air-temperature observation describes the air at the instrument’s location under the observing arrangement used there. A satellite-derived surface temperature describes the radiating surface seen from above. Snow and the air over it need not have identical temperatures. This distinction is the key to reading the Antarctic findings: a lower surface number is important in its own right, but it does not automatically become a new value in a record category based on station air observations.

The familiar Vostok figure of −89.2°C belongs to July 21, 1983. The newer surface estimates belong to remote sensing and subsequent analysis. Compare their dates and measurement types before arranging the values in a single list from coldest to warmest. It also avoids suggesting that an instrument at Vostok simply began reading the satellite value.

The original search covered a long stretch of the plateau

The study described in the original report examined the high East Antarctic ice divide, including the ridge between Dome Argus and Dome Fuji. Its preliminary cold spots were spread along roughly a thousand kilometres. The important finding was therefore not just one unusually cold pixel. The researchers saw a repeated spatial pattern that called for an explanation.

Ted Scambos and colleagues used the different strengths of several satellite records. The older AVHRR series supplied a broad, long-term view; MODIS observations helped identify the cold pattern; Landsat 8 supplied finer spatial detail during the 2013 work. This combination illustrates why a scientific result may depend on more than the newest sensor. A long record and a detailed view answer complementary questions about where an event occurs and whether its pattern recurs.

A landscape that looks flat can contain important hollows

The original report described shallow depressions only a few metres deep within a much larger ice-sheet landscape. A photograph dominated by white snow may make these variations hard to appreciate. Yet a small hollow can matter to air close to the surface even when it is barely noticeable against the scale of the plateau.

The account described dense cold air moving down from slightly higher ground and gathering in these lower areas. Under favourable conditions, that air can remain rather than being quickly replaced. The geometry is local: it is the relationship between a ridge, its slope and a nearby depression that matters, not simply the fact that the place lies somewhere in Antarctica. This helps explain why elevation contours were useful alongside the temperature observations.

Why the dark, clear winter setting matters

The original researchers connected the extreme cooling with prolonged winter darkness and clear atmospheric conditions. A surface can lose energy by radiation even when it is already very cold. The absence of incoming sunlight over the polar night changes the balance compared with an ordinary short winter night at lower latitudes.

That is different from saying every Antarctic location reaches the same minimum. The plateau contains different surface shapes and experiences changing weather. The cold pockets required a particular combination of circumstances, and the report treated those circumstances as something to investigate. A striking low temperature is therefore an observation within a physical setting, not a description of a continent’s everyday weather.

The cracks supplied an unexpected research question

The original search grew out of work on unusual surface cracks. Scambos considered whether intense winter cooling and contraction of the upper snow layer might help explain them. That prompted a closer look at the temperature range and at the coldest areas visible in satellite observations.

This part of the story is easy to lose in a record headline. The researchers were not merely searching for a place to label the winner of a cold contest. A feature of the landscape raised a question about the processes shaping it. The temperature analysis offered a way to investigate that question over a region far larger than a field team could inspect point by point. The crack explanation was presented as a research interpretation, not as a measurement of the age of every visible fissure.

How to read the original temperature map

The map combines surface-temperature observations with information about terrain and coverage. Grey shading represents a compilation of low MODIS readings, with darker areas indicating colder values. Purple squares identify the more detailed Landsat 8 scenes. Green elevation lines provide the topographic context, while the continental outline locates the study within Antarctica. Reading these layers together shows why the cold spots were investigated as a geographical pattern.

These layers should not be read as one instantaneous photograph of a single cold night. They combine observations and geographic information to show a pattern. The red locations distinguish notable temperature sites, including the historical air-temperature record. The full caption accompanying the Ted Scambos / NSIDC image explains how to interpret those locations and layers.

What the 2018 analysis added

NSIDC’s June 2018 account revised the earlier approximately −93°C surface estimates to around −98°C and emphasised the importance of exceptionally dry air. The researchers analysed winter satellite observations from 2004–2016. Their update kept the distinction between the snow surface and the air above it; the result did not turn a remotely sensed surface estimate into a station thermometer reading.

The updated estimate reflects improved data and analysis of extreme surface temperatures. It is not a measurement of the entire plateau becoming five degrees colder between two publication dates. The numbers concern the coldest conditions identified in the observations, while the dates mark different stages in the investigation and interpretation of those data.

Why measurements on the ground remain useful

The original account ended with the challenge of developing weather instruments able to work through the extreme winter. That practical detail matters: a place can be visible to a satellite yet remain difficult to observe continuously from the ground. Electronics, power and the survival of the equipment are part of the measurement problem.

Surface and air measurements made together would help researchers compare quantities that a broad headline tends to merge. The most useful way to read any future cold record is to ask what was measured, by which instrument, at what location and over which period. Those questions preserve the fascination of the Antarctic extremes while giving the figures their proper scientific meaning.

A repeated minimum raised another question

In the preliminary account, many separated cold spots approached similar minimum values rather than each producing a completely different extreme. Scambos treated that repetition as a clue to a possible limit on further cooling under the observed conditions. The question was about the balance between heat leaving the surface and the atmosphere’s influence on that exchange. It was not a claim that nature contains an exact thermostat set to one immutable number.

A hypothesis about the processes limiting further cooling can remain useful even when improved analysis changes the estimated minimum temperature. Better instruments or processing can refine a number without removing the physical question behind it. In this case, both the extreme values and their distribution help researchers examine how radiation, atmospheric conditions and the shape of the snow surface interact during the Antarctic winter.