Intro  |  About Climate Change  |  Consequences  |  Emissions Targets  |  Personal Action  |  Political Action

Temperature changes in March

This is an in-depth look at how the UK climate projections (2009) can be used to understand how climate change will effect temperatures, in Reading, in the month of March. I discuss probability in terms of dice rolling, and introduce you to some real temperature data from the University of Reading Atmospheric Observatory, to help visualise what day-to-day temperatures might be like. It seems likely that, even by the 2050s, and even under lower emission scenarios, we will see March temperatures that are more like the averages that we are used to for April in Reading.

Climate projections for the UK

Having decided the time periods that I am interested in, and that I need to look at both the highest (A1FI) and the lowest (B1) emission scenarios, I am just about ready to delve into the piles of UKCP09 data. The climate projections have been focused in on 25 kilometre squares: you can see from the adjacent map the square that is most appropriate for Reading. Before starting though, I need to decide what probability levels are appropriate. Chapter 2 of the Probabilistic Projections Report talks about this in depth. To try and keep this a tangible and real as I can though, I am going to think in terms of dice...

Temperature changes in March

The measurement we are particularly interested in is called the Dry bulb temperature - this is the normal air temperature, which is recorded in a Stevenson Screen, about 1.5 metres above the ground. A couple of examples of temperatures for the month of March can be seen in the following graphs of March 2003 and March 2009.

• You can see the diurnal (i.e. 24 hour) cycle, with minimum temperatures occurring at night and maximum temperatures in the day.
• There are regular patterns, for example between 16^th and 21^st March 2009 cold nights follow warm days - this may be due to clear skies. At other times, the variation between max and min temperatures is smaller, which may be due to cloud cover.
• There are also breaks in the regular patterns, e.g. 7^th March 2003, which may be evidence that a weather front has passed over.
• For all of this detail though, you will notice that I have drawn two lines on each graph: the average daily maximum and average daily minimum temperatures. It is important to note that the information that we have about future temperatures is only at best an estimate of how these averages will change!
• As a comparison you can see that March 2003 was a warmer month than in 2009. Both average night minimum and day maximum temperatures were about 1.5ºC warmer. The hottest day in 2003 was a good 4ºC warmer, and there were more warm days; there was little difference looking at the coldest nights.

Please notice that the March and August data on this page were downloaded in February 2010. I am using them for illustrative purposes only. The copyright is retained by the Department of Meteorology at the University of Reading.

Long term averages & Comparison between years

The following graph shows some summary data from March 1968 to March 2009. The red dots are the average daily maximum temperatures, which are equivalent to the orange lines in the 2003 & 2009 charts above. The blue dots match the daily / night-time average minimum temperature. You can also see the hottest and coldest temperatures marked for each month.

• Notice that there is a fair amount of variation between years, for example average daily maximums vary by more than 5ºC between years, whereas hottest days vary by over 10ºC.
• You can see that the example years 2003 & 2009 have relatively similar averages, as well as hottest & coldest days. And that they are both relatively warm compared with other years in the dataset.
• I have drawn long term average period lines on the graph. There are three sets of 30 year periods: the first 30 years, the last 30 years and the 1971-2000 period. Notice that this last case can be considered to represent the 1980s as this is the middle decade. The UK climate projections refer to future periods in a similar way, so if I want to look at data for the 2050s, I am actually looking at estimated long term averages for the 2041-2060 period.

Projections for March

The baseline long term average period used by the UK Climate Projections is the thirty years from 1961-1990 (i.e. the "1970s"). For this period the 25 km square that I am using has a March average daily maximum of 9.8ºC, and average daily maximum of 1.9ºC. It is important to note that these averages may be slightly different than for the weather station at the University of Reading.

Visualisation

The following two boxes give you a picture of March in the 2050s and in the 2080s. You can see two blue bars for the Low emission scenario, and red bars for the high scenario. In each case, the one at the top represents where the long term average daily maximum temperature would be, and the one at the bottom represents the long term average daily minimum. Notice that they have been drawn with a thickness to represent the high and low ends of their 67% probability bands. Data is summarised underneath each graph.

• You can hopefully use your imagination to picture what temperatures are going to be like in these periods. Notice that the human mind is actually very good at thinking about and recognising patterns (probably better than most computers). You will need to picture the average day maximum and minimum bands moving up into the range of positions suggested by the low or high scenarios.
• It is important to consider that the example, gray trace of data is simply March 2009 (remember maximum temperatures were a couple of degrees warmer than recent long term average; minimums were close to average).
• How will March averages look from year-to-year? And then how will temperatures vary from day-to-day? Will the amount of variation increase, decrease or stay similar? These are questions that cannot be answered with the information that I have presented...and this is starting to move towards weather predictions, which is beyond the capability of climate models. You can try to guess though - take another look at March 2003 to consider a trace with a wider variation. Maybe think about the estimated changes to the hottest day and coldest night temperatures, from the analysis in the previous section, to help you think about how wide the trace might be.
• At the University of Reading, the April 1980-2009 average daily maximum was 13.4ºC, and average daily minimum was 4.7ºC. Even the bands of the Low emission scenario in the 2050s encompass these temperatures. These figures are in the mid to low parts of the High emission scenario, by the 2080s. It seems quite likely therefore that March temperatures are going to be a lot more like the April temperatures that we have been used to.