Did you know? On a school weather chart, a rainfall bar and a temperature line can cross at exactly the same height without the rainfall and temperature being equal. One is measured in millimetres, the other in degrees Celsius. The drawing brings two stories together, but it does not turn two different measurements into one number.
A dual-axis rainfall-and-temperature chart shows two weather variables over the same periods using separate vertical scales. In a common arrangement, rainfall totals appear as bars read against the left axis in millimetres (mm), while mean temperatures appear as a line read against the right axis in degrees Celsius (°C). Some charts reverse the sides: always follow the legend and axis titles, not the visual convention. To read the graph correctly, identify the mark, choose its named axis, read its unit, then state the relevant month.
One calendar, two measurements, two kinds of summary
A rainfall bar for a month usually represents the total measured rainfall over that calendar month, in millimetres. A monthly mean temperature is an average of temperature observations over that period, expressed in degrees Celsius. Both share the horizontal time axis, but they summarise very different things. A tall rainfall bar does not mean a ‘high average rainfall temperature’ or a rainy day on every day of that month.
Imagine showing a pupil such a graph and asking, ‘Which month was highest?’ The question is incomplete. Highest rainfall? Highest mean temperature? Greatest change from the preceding month? Clarity about the quantity is not fussiness: it is the first step of mathematical literacy.
Worked example: three fictional months, carefully read
This is a hand-constructed classroom illustration, not Singapore’s actual weather or an official climate normal. A school club builds a demonstration chart with these three pairs of values: January: 120 mm rainfall and 28.0 °C mean temperature; February: 80 mm rainfall and 29.0 °C mean temperature; March: 160 mm rainfall and 27.5 °C mean temperature. Each name is a horizontal-axis category. Bars use a left axis labelled ‘Monthly rainfall total (mm)’ and a line uses a right axis labelled ‘Monthly mean temperature (°C)’.
Question one: which month is wettest?
Read only the rainfall bars against the millimetres axis. March has 160 mm, compared with January’s 120 mm and February’s 80 mm. March has the highest monthly total of the three. You have not proved that March had the most rainy days: one heavy storm can affect a monthly total differently from many light showers.
Question two: which month is warmest on average?
Read only the temperature line against the degrees Celsius axis. February has 29.0 °C, compared with January’s 28.0 °C and March’s 27.5 °C. February has the highest displayed monthly mean. That does not tell you the highest single-day temperature. Mean and maximum are different statistics.
Question three: how much did rainfall change?
January to February: 80 mm − 120 mm = −40 mm, a decrease of 40 mm. Relative to January, the percentage change is (80 − 120) ÷ 120 × 100% = −33⅓%, about 33.3% less rainfall than January. This percentage uses January as its base. February to March is a different comparison: (160 − 80) ÷ 80 × 100% = +100%, or twice February’s rainfall total.
Question four: how much did mean temperature change?
January to February: 29.0 °C − 28.0 °C = +1.0 °C. February to March: 27.5 °C − 29.0 °C = −1.5 °C. Do not describe 1 °C as ‘1% hotter’. Celsius degrees on an offset scale do not support that direct percentage interpretation. State a change in degrees Celsius instead.
A short independent arithmetic check
The three listed rainfall totals sum to 120 + 80 + 160 = 360 mm across the three illustrated months. That is not an annual total. Check that the two consecutive rainfall changes reconcile: −40 mm + 80 mm = +40 mm, equal to March’s 160 mm minus January’s 120 mm. The calculation works because the same rainfall unit and chronological endpoints were used.
Why crossings, slopes and heights can deceive
Suppose the February rainfall bar reaches the same vertical height as the temperature line. With separate vertical scales, this is not an equality. An 80 mm rainfall value and a 29.0 °C temperature value can be drawn at the same pixel height simply because the chart designer chose two separate scale ranges. Changing only the right-axis minimum or maximum could move the temperature line while leaving the underlying weather data unchanged.
Likewise, the line may look very steep if the temperature axis spans only 27–30 °C, while the bars may look relatively flat if the rainfall axis spans 0–400 mm. The perceived steepness depends partly on axis scale. To make a numerical claim, read the endpoints and calculate differences in their own units. For more on this visual trap, see how different vertical-axis scales alter the appearance of a comparison.
A useful alternative for younger learners is to draw two vertically stacked charts with the same month labels, one for rainfall and one for mean temperature. That keeps the time comparison while making it harder to mistake a crossing or a bar height for a physical relationship.
An immediate learner diagnostic
Give a pupil the three-month dataset and ask: ‘In February, was the rainfall 29 or 80? Was the mean temperature 29 or 80? Which axis and unit tell you?’ A learner who chooses 29 mm simply because the dot sits above a bar is reading position without following the variable. A learner who identifies 80 mm rainfall and 29.0 °C mean temperature has connected marks, axis and units.
Next ask whether the chart proves rainfall caused the temperature change. The answer is no. Even if two series rise or fall together in a short display, the chart by itself cannot isolate the cause, separate seasonal influences or establish a general relationship. One three-month constructed example is especially unsuitable for a climate conclusion.
Practise from guided reading to independent interpretation
First exercise: annotate the chart
Using two colours, trace the rainfall bars to their correct axis and circle ‘mm’. Trace the temperature line to its axis and circle ‘°C’. For each of the three months, write a paired sentence, such as ‘January had 120 mm of rainfall and a mean temperature of 28.0 °C.’ Correct unit labels matter as much as the numerical answer.
Second exercise: calculate without borrowing the wrong scale
With a partner, compare February and March rainfall using millimetres, then compare their temperature means using degrees Celsius. Swap papers and check the signs: rainfall increased by 80 mm and temperature mean decreased by 1.5 °C. Explain why the answers cannot be combined into one ‘total weather change’.
Independent transfer: redesign a misleading graph
Provide an unfamiliar chart where temperature occupies the left axis and rainfall the right, with the colours swapped. Ask students to write three true sentences, one unsupported inference, and a clearer redesign. Their conclusions must work from the data and legends rather than from the earlier chart’s layout.
For an advanced extension, ask whether averaging the three monthly mean temperatures gives the exact mean temperature over all calendar days in the period. Not necessarily: calendar months have different numbers of days and may contain different numbers or timings of observations. A whole-period mean needs a documented weighting and sampling method; the average of three monthly means is not automatically the same thing.
What can go wrong in a real weather dataset?
- Missing data: a monthly rainfall total may be incomplete if a gauge or reporting record is missing. Do not silently treat missing rainfall as zero.
- Different locations: rainfall recorded at one site and temperature measured at another may not describe identical conditions.
- Different periods: do not compare a monthly rainfall total with a daily temperature maximum as though both summarise the same interval.
- Different climate summaries: distinguish measurements in one named year from multi-year climate averages or normals.
- Unclear chart conventions: check whether symbols use the left or right axis; colour alone is not an accessible substitute for labels.
- Insufficient evidence: a short chart can describe an observed pattern but cannot on its own establish cause, a long-term climate trend, or next month’s weather.
The US National Weather Service’s chart explanation resource illustrates why weather displays require careful attention to the variables and their plotting conventions. For authoritative climate analysis, use properly described observations rather than pretending these invented classroom readings are official records.
A progress check parents and teachers can actually use
Start with an unlabelled explanation from the learner: ‘The blue line is higher, so there was more rain.’ After teaching, give them a second chart with the two axes reversed. Can they identify each unit, correctly retrieve four values, calculate two changes and flag one conclusion the chart cannot support? An improvement from visual guessing to correctly sourced values is observable transfer.
Parents can ask a child reading a weather infographic to point to the axis rather than asking only for the answer. Teachers can require a short ‘variable, unit, period, statistic’ note next to each plotted claim. Tutors can diagnose the specific bottleneck—legend reading, axis mapping, units, subtraction or causal overreach—before assigning another worksheet. This reflects the approach in the eduKate Clementi small-group Mathematics reference: explain the misconception, model a check, then see whether the learner can do it independently.
Frequently asked questions
Can a rainfall bar touch the temperature line without anything unusual happening in the weather?
Yes. The apparent intersection is a graphical coincidence under chosen vertical-axis scales. It does not represent an equality between millimetres and degrees Celsius.
Is the month with the most rainfall the month with the most rainy days?
Not necessarily. The rainfall total sums the measured precipitation depth for the period, while a rainy-day count depends on a specified daily rainfall criterion. The variables answer different questions.
Can I use a temperature anomaly as though it were actual temperature?
No. A temperature anomaly is a difference from a stated reference baseline; the number is not the same as the observed absolute temperature. Identify what the particular chart labels and see how temperature anomalies differ from actual temperatures.
Should every school climate chart use two vertical axes?
No. Two separate panels may be easier to read and less likely to suggest a false comparison. A dual-axis presentation is useful only when clearly labelled and carefully interpreted.
What this teaches beyond weather
Reading a two-axis chart is a lesson in disciplined attention. The learner learns not to let one picture erase the distinction between two measurements. This transfers to budget dashboards, health charts, school science results and public statistics: find the variable, unit, denominator, time window and evidence before announcing a conclusion. For more connected learning, explore the How Education Works hub.