Chapter 12 of 17 · Water and the Weather
Chapter 11. Weather and climate
Two words that are confused every day, the scale of time that separates them, and what a climatological normal is.
Two words, two scales of time
Nobody confuses a day with a lifetime, and yet many people confuse the weather with the climate. Both speak of the same quantities, are measured with the same instruments and use the same words. What separates them is neither the object nor the method: it is the scale of time over which one looks.
The weather describes what it is like now, here, or in the days to come. It is measured in hours and days. The previous chapter was devoted to it entirely. The climate describes the usual behaviour of the atmosphere in a place, over tens of years. It is not measured in a day or in a month, and no particular day tells us anything about it.
This difference of scale is not a specialist's nuance. It decides what one is entitled to conclude. A very cold morning says nothing about the climate, and a warming climate stops no morning from being very cold. This chapter installs the distinction and gives the tool that makes it possible to hold both ideas together without contradiction.
The objectives of this chapter
Each one is observable. By the end of the chapter, either you do it, or you know which section to read again.
- Tell a statement about the weather apart from a statement about the climate by its scale of time.
- Define a climatological normal, and say over what length of time it is calculated.
- Name the reference period used in France today, and say how often it changes.
- Explain why a particular day allows no conclusion about the climate.
- Read a departure from the normal, and say what it does and does not allow you to conclude.
What this chapter assumes is already known
Two prerequisites, installed by earlier chapters of this volume: the quantities and the conditions of measurement on the one hand, the making of a forecast on the other.
The climatological normal, and why thirty years
To say that a month has been warm, one has to know warm compared with what. That is the role of climatological normals: statistical references that represent the climate of a given period, and against which what is measured today is compared.
These normals are calculated over periods of thirty years, and they are updated every decade. This rule comes from the World Meteorological Organization, and national services follow it. Thirty years is not an arbitrary number: it is long enough for exceptional years to cancel one another out, and short enough to stay representative of a period.
Météo-France changed its reference period on 28 June 2022, moving from normals calculated over 1981-2010 to normals calculated over 1991-2020. The reason given is explicit: the 1981-2010 normals were no longer representative of the present climate in a context of climate change. The service states that the new normal puts the mean temperature of France at about 13 degrees, that is 0.42 degrees more than the previous reference period.
This last fact deserves pausing on, because it is instructive about what a normal is and is not. A normal is not a fixed value of nature: it is an average calculated over a chosen period, and it moves when the period moves. To compare a year with the normal is therefore to compare it with a dated reference, and one has to know which one.
The words to remember
Four words. The first two are the ones people confuse, the next two are the ones that make it possible to stop confusing them.
- Meteorology
- The study and forecasting of the state of the atmosphere in the short term: what it is like today, what it will be like in the days to come. Its scale is the hour and the day.
- Climate
- The usual behaviour of the atmosphere in a place, described by averages, by the spread around those averages, and by extreme values. Its scale is the decade. No particular day tells us anything about it.
- A climatological normal
- A statistical reference calculated over a period of thirty years, updated every decade, and against which measured values are compared. The period used in France today is 1991-2020.
- A departure from the normal
- The difference between a measured value and the corresponding normal. It can only be read by knowing over which period the normal was calculated: the same measurement gives two different departures according to the reference used.
Recognising a statement about the weather and a statement about the climate
Statements about the weather
- It will rain tomorrow afternoon over the northern half of the country.
- The temperature dropped to minus three degrees last night.
- 18 mm fell in three hours on the district.
- The wind will reach 70 km/h in gusts at the end of the day.
- This week has been the coldest since the start of the winter.
Statements about the climate
- July is on average the warmest month of the year in this region.
- The mean annual temperature of France stands at around 13 degrees over 1991-2020.
- The precipitation on this slope is heavier than on the opposite slope.
- The reference period used today is 0.42 degrees warmer than the previous one.
- This region has on average few days of frost.
What one day never proves
Here is the commonest mistake of reasoning on this subject, and it is made in both directions. The first direction: an exceptional cold spell is taken as proof that the climate is not warming. The second: a heatwave day is taken as proof that it is.
Both pieces of reasoning have exactly the same flaw: they draw a conclusion on a scale of tens of years from an observation made over a single day. Since climate is defined by averages over thirty years, no single value can either confirm or contradict it. That is true by construction, not out of caution.
A comparison helps to feel why. Knowing the mean height of a group of a thousand people does not allow one to guess the height of the first person one meets; and meeting a very tall person changes nothing about the average of the group. Both pieces of information are true, they bear on different objects, and one never refutes the other.
What tells us about the climate is long series: measurements made in the same place, under the same conditions, over decades. That is exactly why the previous chapter insisted so much on standard conditions. A series whose conditions changed along the way allows no comparison over time, and a series that cannot be compared says nothing about the climate.
So remember the logical order of this whole volume, which closes here: we measure under shared conditions, we obtain comparable series, we draw normals from them, and only then can we speak of climate. There is no short cut.
The trap of this chapter: believing that a normal is a forecast, or that it describes what ought to happen. A normal is a past average, nothing more. A month that departs from its normal is not an abnormal month in the sense of being at fault: departures are the rule, not the exception, and an average is precisely what real values depart from. What tells us something is not the existence of a departure, it is the repetition of departures in the same direction over decades. A single departure is noise; a trend is a signal.
Worked example. Sorting three statements
Question: three sentences are put forward. One, there was frost yesterday morning in the valley. Two, this region has on average forty days of frost a year. Three, the current reference period is warmer than the previous one. Sort each one as weather or as climate, and say what each one allows you to conclude. Example guided up to the next-to-last step.
- Set out the sorting criterion
The criterion is the scale of time the statement bears on, and it alone. A statement bearing on one day or on a few days is weather. A statement bearing on tens of years, or on an average calculated over a long period, is climate. - Sort the first
There was frost yesterday morning in the valley: that bears on one night. It is a statement about the weather. It allows one to conclude that there was frost that night, and strictly nothing else. - Sort the second
This region has on average forty days of frost a year: the words on average and the unit a year point to a statistic over many years. It is a statement about the climate. It allows one to say what to expect in general, and not what will happen this particular winter. - Sort the third
The current reference period is warmer than the previous one: that compares two periods of thirty years. It is a statement about the climate, and it is even the strongest form such a statement can take, since it compares two long series with each other. - Over to you: find what the first does not prove
Somebody uses the first statement to dispute the third. Explain in two sentences why the reasoning does not hold, using the words scale of time and series. Check it in the answer to the self-check, question 5.
We measure under shared conditions, we obtain comparable series, we draw normals from them, and only then can we speak of climate.
What to remember from this chapter
- Weather and climate bear on the same quantities: what separates them is the scale of time, the hour and the day on one side, the decade on the other.
- A climatological normal is a statistical reference calculated over thirty years and updated every decade, following the rule of the World Meteorological Organization.
- Météo-France has used the period 1991-2020 since 28 June 2022, in place of 1981-2010, which had become unrepresentative of the present climate.
- On that new reference, the mean temperature of France is about 13 degrees, that is 0.42 degrees more than over the previous period.
- No particular day either confirms or contradicts a statement about the climate: that is true by construction, since climate is defined by long averages.
- A series whose conditions of measurement have changed says nothing about the climate: the standardisation of chapter 9 is what makes this chapter possible.
Self-check for chapter 11
Answer in writing, using only the knowledge given by this volume. The answers are gathered at the end of the volume.
- Give the single criterion that allows a statement to be sorted between weather and climate, then apply it to two examples of your own choosing.
- What is a climatological normal, over what length of time is it calculated, and how often is it updated?
- Why did Météo-France replace the period 1981-2010 with the period 1991-2020? What becomes of a departure from the normal calculated on the old reference?
- Explain why chapter 9 was necessary for chapter 11 to be able to exist.
- Somebody states that an exceptional night of frost contradicts the idea of a warming climate. Refute that reasoning in two sentences, using the words scale of time and series.