the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Mar 2026
The Cold Spell That Never Happened: Climatological Assessment of the Winter 2025/26 Cold and Snowfall Episodes in Bucharest and Their Disproportionate Media Representation
Abstract. The winter of 2025/26 in Romania generated intense media coverage characterizing cold and snowfall episodes in Bucharest and south-eastern Romania as historically exceptional. Here, we evaluate this characterization against a multi-source climatological analysis spanning up to 148 years of instrumental records. Using daily minimum temperature data from Bucuresti Filaret station (1879–2026) and monthly mean temperatures from 65 stations across Romania, we show that both January and February 2026 were anomalously warm relative to the 1971–2000 baseline (country-mean anomalies of +2.3 °C and +2.2 °C, respectively), with positive temperature anomalies recorded at nearly all 65 stations. Cold spell detection, using the 10th percentile threshold of the daily minimum temperature (TN10p), confirmed zero cold spell days throughout the winter 2025/26. The principal high-impact event, the 18th of February 2026 blizzard, was driven by a classical Mediterranean cyclone characterized by an upper-level potential vorticity streamer and an anomalously high integrated water vapor transport directed towards south-eastern Romania. Content analysis of 89 Romanian and international media items (~112,400 words) reveals 692 alarm-vocabulary occurrences (6.2 per 1,000 words). Outlet alarm-vocabulary density correlates strongly and negatively with the provision of historical context (Pearson r = −0.88; p < 0.001). We interpret this amplification as consistent with both the Social Amplification of Risk Framework and a shifting experiential baseline mechanism. Our study demonstrates that heavy snowfall in a warming climate can occur in the complete absence of any thermal cold extreme, and underscores the practical consequences of conflating precipitation and temperature hazards in emergency communication, and proposes a standardized context protocol for national meteorological services.
- Preprint
(8100 KB) - Metadata XML
-
Supplement
(3141 KB) - BibTeX
- EndNote
Status: open (until 20 Aug 2026)
- CC1: 'Comment on egusphere-2026-1457', Ion Marinică, 23 Apr 2026 reply
-
RC1: 'Comment on egusphere-2026-1457', Ion Marinică, 25 Apr 2026
reply
2026.04.24
In Romania, the Blizzard is well studied. From the classical specialists of old (e.g.: Ștefan Herpites) to the modern ones (e.g.: Ion Drăghici) who explained very well the synoptic and physical processes that occur during the Blizzard situation. In general, for Romania and the Balkan Peninsula there are two main types of Blizzard:
I. ) when the Cyclone-Anticyclone coupling is between a Mediterranean Cyclone (or Black Sea) and the East-European Anticyclone. This type of Blizzard is also called by specialists as "Eastern Blizzard" because the advection of cold air is made from the Eastern sector (from the East and Northeast). During the Blizzard, the air temperature is usually not very low but the wind intensification accentuates the feeling of cold, etc. Such a Blizzard is always followed by an intense cooling of the weather (cold wave or frost wave). It is the most common type of Blizzard.II. ) When the Cyclone-Anticyclone coupling is made between a Mediterranean Cyclone and an Atlantic Anticyclone or the ridge of such an anticyclone which can be Azoric or the ridge of a Greenland anticyclone and sometimes even Scandinavian, etc. (I will not go into details.) In this situation, the snowfalls are quantitatively important in the rear sector of the cyclone, and their diminution or cessation is made quite quickly. Such a Blizzard is also called by specialists as "Western Blizzard" because the advection of the cold air mass is made from the Western sector (West or frequently Northwest). It is known that generally after snowfalls, the weather cools down. It is also known that western blizzards are less frequent and the cooling that follows after the cessation or decrease of snowfall is not intense (due to the influence of the Atlantic Ocean). (I will not go into details.). In addition to these two main types, there are other blizzard situations without major importance.
Therefore, the cooling that followed the Blizzard was moderate and the forecast prepared by the ANM (Romania's elite institution) did not have any reference to an intense cooling of the weather or the appearance of a cold wave. The lowest thermal minimums were recorded on the morning of 19.II.2026 and from these I quote (excluding the mountain area): -13.0°C in Bacău (in Moldova), -9.6 °C in Ploiești in Muntenia), -5.0°C in Băneasa (in the Capital area), etc. The following day (20.II.2026) the minimums increased.
Although the scientific work of the authors Monica Ionita, Bogdan Antonescu, and Viorica Nagavciuc (The Cold Spell That Never Happened: Climatological Assessment of the Winter 2025/26 Cold and Snowfall Episodes in Bucharest and Their Disproportionate Media Representation) does not aim to provide a synoptic analysis of this winter's blizzard but only the amplifications made by the media, I added this comment to highlight that often various non-specialists are invited by television stations to make some comments that are usually amplifications of past, present or future atmospheric phenomena and processes. Most of the time, real specialists in the field of meteorology are not asked anything, as if they did not exist.
Additionally, in the last 25 years, the political factor has infiltrated climate science and as a result almost no one does science anymore but only politics (usually of the globalist type).
If politics enters science, the latter disappears and everyone does politics with the scent of science! To have science persevere, free of political commitment, it has to be based on true fundamental logic (binary mathematical logic, the true and correct one).It is important to protect and save man, because "the planet saves us and not the other way around".
I note again that the publication of this work is welcome.
Citation: https://doi.org/10.5194/egusphere-2026-1457-RC1 -
RC2: 'Comment on egusphere-2026-1457', Anonymous Referee #2, 29 Apr 2026
reply
A question regarding the text search process. The authors clearly show how the 18th Feb 2026 event is not an extreme cold event and that the resulting snow accumulation is not historically unprecedented for Romania. Thus, any claim of such by the media would be inaccurate. In lines 436-437, it is stated “'unprecedented' appeared 40 times despite being factually incorrect for every observable metric of the 2026 winter”. However, could it not be argued that, as a single day snowfall event in Bucharest specifically, it is unprecedented, e.g. lines 389-391? Is there any screening of the search results to identify this, admittedly specific, example?
-
RC3: 'Reply on RC2', Ion Marinică, 30 Apr 2026
reply
30.IV.2026
Heavy snowfalls in a single day are quite common events (just as torrential rains are quite common), in winter and possible anywhere on Earth. Therefore, we cannot say that "heavy snowfall in a single day" is an exceptional meteorological event.
Citation: https://doi.org/10.5194/egusphere-2026-1457-RC3
-
RC3: 'Reply on RC2', Ion Marinică, 30 Apr 2026
reply
-
RC4: 'Comment on egusphere-2026-1457', Joanna Wibig, 14 Aug 2026
reply
The paper concerns a case of winter 2025/2026 in Bucharest in Romania, which generated intense media reports about historically exceptional cold and snowy weather. The authors confront temperature and snow fall in Bucharest and Romania in January and February 2026 with mean values in the reference period 1971-2000 and with historical data from five stations in Romania from the period starting at the end of the nineteenth century. They have shown that mean monthly temperature and mean monthly minimum temperature are characterized by a positive anomaly when compared with mean values from the reference period. But in case of snowfall episodes the one from 18 February from Bucharest was the highest obseverd in the record starting at the end of the 19th century, however both the duration and the maximum depth of the snow cover was far from the extreme values. Than they analyse the reports of media items Romanian and international and indicate that the media representations diverge systematically from the scientific, report an exceptionally cold and snowy winter, although in reality it was warmer than typical in the reference period, and although a single snowfall was indeed unprecendented, the maximum snow cover thickness was significantly lower than typical for the reference period.
This is an interesting study, but several elements raise my serious reservations..
The introduction should include a presentation of the problem and a characterization of similar cases from the past or from other regions of the world. However, in this work, the introduction summarizes the research findings, quite well indeed.
The analysis of thermal and snow conditions in January and February 2026 against the background of the reference period (i.e. 1971-2000) was performed very precisely and raises no objections.
However, the unique nature of the events is problematic. In the case of mean daily temperature and mean daily minimum temperature, January and February 2026 were truly warm compared to the reference period. However, as the authors themselves point out, the average person relates weather conditions to their recent experiences, i.e., the period of the last few years. We typically wear clothing and footwear appropriate for such weather conditions, not those appropriate for the weather during the reference period. In the case of rapidly progressing warming, the differences are very pronounced.
In the case of snowfall, the single-day snowfall of 45 cm, as the authors themselves admit, was the highest recorded in Bucharest since the beginning of the measurement series, i.e., the end of the 19th century. The fact that the snow cover was significantly lower than previously recorded does not diminish the exceptional nature of this event. The authors themselves mention serious traffic disruptions and power outages for over 200,000 people. Regular snow removal after several centimeters of snowfall is one thing, but the need to quickly clear 45 cm of snow in the absence of electricity is quite another. Additional problems include defrosting refrigerators, spoiling vast quantities of food, the inability to prepare hot drinks and meals, and often even lack of heating and water. Add to this the difficulty of refueling or shopping. All those who personally had to deal with this undoubtedly consider this event extreme and/or catastrophic.
Another problem is the analysis of media reports. Firstly, most of it, including all tables and figures, was moved to the supplement. Secondly, the analysis was not separated into the sections concerning reports on thermal conditions, which were indeed overly dramatic, and the reports on snowfall on February 18, 2026. Only a few days in January were cold, and calling them extremely cold is a gross exaggeration, but the snowfall was indeed extreme.
The recommendations proposed by the authors are also unclear. Who are they aimed at? Meteorological services should be responsible for providing accurate information about weather conditions and the resulting threats to the public. Administrative authorities and services responsible for public health and safety should obtain precise information and possible warnings from them, not from the media. Analyzing the uniqueness of events should also be a task for the meteorological service. It's unreasonable to expect journalists to be familiar with climate series dating back several decades. From my own experience, I know that journalists readily use such analyses if they are available.
Other notes regarding the recommendations:
Due to the rapid pace of climate change, it has recently been recommended to choose the last thirty years ending with a full decade as the reference period (in this case, the years 1991-2020). These are still values within the grasp of the average reader.
Journalists should use language understandable to the average reader, so the percentile rank of the anomalies and current TN10p values are generally inadvisable. The multitude of definitions of coldspell would require specifying the current definition, which complicates matters.
Generally, the recommendations contained in this text are recommendations for meteorological service announcements rather than media reports.
However, I fully support the idea that the media should undertake targeted educational initiatives to support improvements in weather and climate literacy, particularly with respect to the interpretation of warnings and the historical context of extreme events as well as presenting concise explanatory material on the interpretation of extreme-weather indicators and terminology, particularly the distinction between anomalous, severe, and climatologicaly extreme conditions. Both should be made in cooperation with local meteorological services or scientists.
Citation: https://doi.org/10.5194/egusphere-2026-1457-RC4
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 985 | 573 | 53 | 1,611 | 197 | 52 | 66 |
- HTML: 985
- PDF: 573
- XML: 53
- Total: 1,611
- Supplement: 197
- BibTeX: 52
- EndNote: 66
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Publisher’s note: this comment is a copy of RC1 and its content was therefore removed on 27 April 2026.