Learn why you need to consider 30 years of data when making decisions concerning future climate change or determining how the climate has changed at a specific location.
When it comes to making decisions which incorporate future climate change, or determining how the climate has changed at a specific location, the advice is to use at least 30 years of data.
Climate varies naturally over many different time scales – from season to season, from year to year, and from one decade to the next. Many of these variations are caused by the interactions and feedbacks between the different components of the climate system – atmosphere, oceans, land and ice – many of which are chaotic, or unpredictable.
Some phenomena, however, occur on more or less regular cycles, the most well-known of which is probably El Niño/ La Niña (every 3-7 years).
There are also other natural factors which can have an effect on our climate – for example, volcanic eruptions can eject vast amounts of particulate matter into the atmosphere which can quickly spread around the globe and impact climate.
In addition to these natural forcing factors, humans have had, and will continue to have, an effect on climate.
Human-induced greenhouse gas emissions are also altering the Earth’s climate by changing the amount of heat trapping gasses in the atmosphere and increasing its greenhouse (or warming) effect. These gases are long-lived and will have an effect on climate for many years to come.
So, how do we know whether a particular year is warmer or colder than average?
How do we know if there is a trend in our climate records, or indeed, what is the average climate?
To get a good idea of our average climate, we need to examine enough data to be sure that we are capturing the influence of as many of these different forcing factors as possible, and not just some of them. For example, we need to make sure that we have included the effect of both El Niño and La Niña events, as they affect climate differently.
Climate normals have two main roles:
To serve as a reference against which conditions can be assessed, e.g., was this winter warmer than average, or how much warmer will it be in the 2050s (2041-2070) compared to the present climate?
To give an indication of the conditions likely to be experienced at a particular location, e.g., I’m planning a holiday to Kelowna – how hot is it likely to be in July?
The World Meteorological Organization considers a thirty-year period to be the minimum required to calculate the average climate, known as a climate normal. Climate normals are updated at the end of every decade.
We can use climate normals to help determine if there are trends in a climate record.
Let’s look at the change in Canada’s annual average temperature from 1948-2018, when compared to the 1961-1990 average. The year to year differences compared to the reference period average are know as anomalies.
If we look at the anomalies in 10-15 year chunks and calculate the trend over that time, you can see that the trend changes – sometimes it is increasing, sometimes it is decreasing and sometimes there is almost no change. If you look at the trend over the whole length of record, however, Canada’s annual average temperature has increased by almost 2°C.
The trend over the shorter time periods gives an indication of the natural variability of climate, while the longer term trend is a result of climate change. Because of natural climate variability, it is possible to experience short-term trends that are opposite to the overall trend due to climate change.
Annual average temperature change for Canada as a whole, with respect to the 1961-1990 climate normal period.
Blue line with markers – year to year anomalies with respect to the 1961-1990 average
Red dashed line – trend in climate over the time period shown
Scientists use models of the climate system to simulate how the climate has and will evolve according to the different forcings. These simulations include natural climate variability, so considering less than 30 years of data may reflect a trend that is different/opposite to that of longer term climate change.
To demonstrate this, consider the output of one of the 24 climate models on ClimateData.ca, the number of days with maximum temperature greater than 30°C.
Looking at the whole graph, from 1950 to 2100, the trend is increasing.
Similarly, looking at a thirty-year period (2041 to 2070), the trend is increasing.
To demonstrate this, consider the output of one of the 24 climate models on ClimateData.ca, the number of days with maximum temperature greater than 30°C.
Looking at the whole graph, from 1950 to 2100, the trend is increasing.
Similarly, looking at a thirty-year period (2041 to 2070), the trend is increasing.
But what if a shorter period was considered? There are a couple of 10-year periods in this time series (blue), where the trend is decreasing.
These 10 year-periods are not representative of the longer-term climate change trend.
Decisions (e.g. risk assessments or adaptation actions) based on the trend of a shorter time period may lead to insufficient actions, inappropriate for the longer-term climate trends.
ClimateData.ca allows you to do this for any of the climate variables and indices available on the Map page, and uses 1971-2000 as the reference period against which future climate is compared.
