Skip to content


Climate change

Air temperature changes until now

In Belgium, over the course of the 20th century, there were very marked and quite severe increases in seasonal and annual temperatures (in the order of 1°C) during two periods, firstly during the first half of the 20th century and then in the 1980s (2). According to a recent report (5), meteorological observations in Ukkel since 1830 point to a temperature increase of some 2 °C over the entire period 1830-2010. The frequency of heat waves shows a significant upward trend towards the middle of the 1990s. However, the variability of this parameter is important throughout the 20th century and the characteristics of heat waves in recent years are relatively similar to those observed in the 1940s, following the warmer summer temperatures observed during the first part of the 20th century (2).

Moreover, the frequency of cold spells reduced significantly in the early 1970s. The general increase in minimal temperatures during the 20th century also explains why the longest annual period without frost has increased. In fact, the last day of frost at the end of winter tends to arrive earlier, with the first sign of frost at the onset of winter now being later (2).

In recent decades, a reduction in the amplitude of daily temperature changes has been observed (night minimums rise faster than day maximums). Increased cloud cover is a very likely contributor to this change and some models suggest that it will continue to increase in the future (4).

On the Ardennes plateau, in the Saint-Hubert region, maximum annual snow cover has shown a very significant fall since winter warming began in the late 1980s (4).

Precipitation changes until now

Between 1833, when rainfall records began, and the end of the 20th century, the Brussels region has seen an increase of roughly 7% in annual rainfall (highly significant) with rises of around 15% in winter (highly significant) and spring (significant). Moreover, over the past 50 years in the country, most climatological stations have revealed a trend towards significant or highly significant increases of annual extremes of rainfall extending over several days; this type of extreme precipitation event usually occurs in winter. On the other hand, annual maxima for precipitation over 24 hours (or for even shorter periods) remain stable, except near the coast where, according to a recent study, daily annual maxima are already showing a significant increase (2).

Annual maxima for precipitation ranging between 1 and several hours do not show any marked changes since 1898, however. In spite of some record values over recent years, no significant change has been detected either in the annual frequency of the number of days where precipitation reached at least 20 mm (2).

From all the available precipitation data, it can be concluded that neither the intensity nor the frequency of violent storms in the Brussels area have shown any marked increase since the start of the 20th century. Using the annual maxima analysis of daily precipitation recorded in the Belgian climatological network, a similar conclusion can be drawn for the whole of the country over the past 50 years, with the exception of the area close to the coast (2).

With regard to drought, a preliminary study undertaken by the Royal Meteorological Institute of Belgium shows that the longest periods without significant precipitation recorded at their station near Brussels reveal no major change since the early 20th century (2).


Hailstorms have the potential to cause substantial damage to hail-susceptible objects such as buildings, crops or automobiles. Prominent examples are the two hailstorms related to the low-pressure system Andreas that occurred on 27- 28 July 2013 over central and southern Germany with total economic losses estimated at approximately EUR 3.6 billion (16). There is no good overview of hail events in Europe, however. Little is known about local hail probability and related hail risk across Europe. The majority of Europe is not covered by a hail network, and this leads to a gap in direct hail observations (15).

For a large part of Western Europe, covering Germany, France, Belgium and Luxembourg, the occurrence of hailstorms was mapped over a 10-year period (2005–2014) (15). The results show a sea-to-continent gradient in the number of hail days per year: an increasing gradient in the number of hail days per year can be recognized from north-western France towards central France, and from northern towards southern Germany. The highest number of severe storms is found on the leeward side of low mountain ranges such as the Massif Central in France and the Swabian Jura in southwest Germany. In this study area and study period, hail day frequency was low over north-western France, Belgium and northern Germany.

Wind climate changes until now

Near Brussels a very marked fall in annual mean wind speed has been recorded during the second part of the 20th century. However, over the years the development of vegetation around the measurement site makes it impossible to attribute with certainty this trend to climate change alone. Elsewhere in the country, in some stations, wind measurements that are probably more reliable for the study of its changing characteristics have been recorded since the 1960s. An analysis of this data indicates a relatively severe reduction in wind speed in the 1980s, with a slight accentuation of this trend since that time (2).

With regard to storms, analyses conducted to date into strong winds, since 1940 in the case of the station near Brussels and since 1985 elsewhere in the country, have revealed no particular trends, either in the intensity of the strongest winds each year or in the frequency of high winds (2). Storm intensity estimates derived from wind, wave and surge observations from Belgium show no trend over the period 1925–2007 (13). 

A northward shift in mean storm track position since about 1950 is consistent in studies on wind climate in northwestern Europe over the last decades (12). This northeast shift together with the trend pattern of decreasing cyclone activity for southern mid- latitudes and increasing trends north of 55 - 60°N after around 1950 seems consistent with scenario simulations to 2100 under increasing greenhouse gas concentrations (14).

Air temperature changes in the 21st century

Temperature projections for Belgium, which are illustrative of the global trend, predict winter temperature increases ranging between 1.5°C and 4.4°C and summer temperature increases between 2.4°C and 7.2°C (up to + 8.9 °C in August) by the end of the 21st century (6).

Future cold spells in Western Europe are projected to become about 5°C warmer (and remain above freezing point), thus having a significant climatic impact. This conclusion is based on research in which a cold spell (CS) is defined as a non-interrupted sequence of days in which the 5-day average temperature falls below a threshold value Tcold (8).

In much of the North Sea region the number of tropical nights at the end of this century may rise by about 10 days under an intermediate (RCP4.5) and by more than 20 days under a high-end (RCP8.5) scenario of climate change, with tropical nights hardly ever occurring in this region under the present-day climate. Similarly, cold spells are projected to become shorter in the North Sea region, by about 3 days for the intermediate (RCP4.5) and about 4-6 days for the high-end (RCP8.5) scenario. Warm spells are projected to become markedly longer in the North Sea region, by about 30 days for the intermediate and by 60-120 days for the high-end scenario (9). 

Precipitation changes in the 21st century

In Belgium, projected changes in winter precipitation during the 21st century reveal a moderate increase in precipitation (between 3 and 30%) (4), whereas summer precipitation is expected to decrease, although quantitative findings vary here (ranging from the status quo to a fall of up to 50%). Others report a regional projected maximum increase in winter precipitation up to 60 % and a maximum decrease in summer precipitation with 70 % for the period 2071-2100 compared to 1961-1990 (7), based on regional and global climate models At the Belgian coast and in the Ardennes region, the precipitation increase is expected to be more important while the precipitation decrease in summer is expected to be less strong than other areas in Belgium (7). In Belgium, the frequency of significant rain events is also set to rise (3).

The total annual amount of precipitation will probably decrease (4).

In their 5th Assessment Report the IPCC presented a precipitation decrease at the end of the 21st century for April through September up to 10% for England, Belgium, the Netherlands and northern Germany, under the intermediate RCP4.5 scenario of climate change. These projected changes, however, do not exceed natural climate variability across the region. For October through March a precipitation increase up to 10% was projected for this North Sea region; these projected changes do exceed natural climate variability across the region (10).

Wind climate changes in the 21st century

Some general circulation (i.e. global) models suggest an increase in the intensity and/or frequency of the strongest storms over Europe, but there is still some debate on the explanation and generality of this result (4).

A review of recent scientific literature shows that the projected changes in wind extremes (speed and direction) for the North Sea region are
typically within the range of natural variability and can even have opposite signs for different scenarios either simulated by different climate models or for different future periods (11). 


The references below are cited in full in a separate map 'References'. Please click here if you are looking for the full references for Belgium.

  1. Van Ypersele and Marbaix (2004)
  2. Royal Meteorological Institute of Belgium (2009), in: Ministry for Social Affairs, Health and Environment (2009)
  3. ADAPT (2008), in: Ministry for Social Affairs, Health and Environment (2009)
  4. Ministry for Social Affairs, Health and Environment (2009)
  5. Royal Meteorological Institute of Belgium (2009), in: National Climate Commission Belgium (2010)
  6. Baguis et al. (2009), in: National Climate Commission Belgium (2010)
  7. CCI-Hydr (2009), in: National Climate Commission Belgium (2010)
  8. De Vries et al. (2012)
  9. Sillmann et al. (2013), in: May et al. (2016)
  10. IPCC (2013), in: May et al. (2016)
  11. May et al. (2016)
  12. Feser et al. (2015a), in: Stendel et al. (2016)
  13. Hossen and Akhter (2015), in: Stendel et al. (2016)
  14. Ulbrich et al. (2009); Feser et al. (2015a), both in: Stendel et al. (2016) 
  15. Fluck et al. (2021)
  16. SwissRe (2014); Kunz et al. (2018), both in: Fluck et al. (2021)

Share this article: