Weather, climate and reading trends in data

1. Climate includes a distribution, not only an average

A weather description might give today’s temperature, wind and rainfall at a particular location. A climate description asks what conditions tend to occur across many years at that time of year. It includes how much observations vary and how often unusual values occur. Two places can share the same mean temperature yet differ in daily range or frequency of very hot days. Averages are summaries of a distribution, not substitutes for the distribution itself.

WMO climatological standard normals use consecutive thirty-year periods, such as 1991–2020, to provide reference summaries. Thirty years is a convention for a useful reference, not a magical boundary below which nothing can be learned. Different questions may require longer records or another baseline. Always compare like seasons: contrasting one place’s winter with another place’s summer would mix the seasonal cycle with geographic differences.

Sources: NCERT: World climate and climate change ↗ · WMO: Climatological normals ↗

2. Build comparable records before drawing a trend

A station record can change because the climate changes, but also because an instrument, observing time or station location changes. A thermometer moved from an open field to a sun-heated wall would not provide a clean continuation of the earlier record. Metadata record such changes. Quality control checks implausible observations; homogenisation investigates artificial breaks using station history and comparisons with other records. Adjustments should have documented methods and uncertainty.

Regional estimates also need sensible spatial weighting. Adding many new stations in a cool mountain area must not automatically make the whole region appear to cool. Scientists combine measurements over defined areas and assess incomplete coverage. Missing data are not zero temperature or zero rain. The absence of a reading is information about the record, not proof of a calm or rainless day.

Sources: NASA: How scientists measure global temperature ↗ · NASA GISS: Temperature anomalies, baselines and data quality ↗

3. Worked case: subtract a matching reference

A temperature anomaly is observed temperature minus the reference mean for the same place and time of year. Suppose a hypothetical coastal station has a July reference of 28°C and a particular July mean of 29.2°C. Its anomaly is +1.2°C. A hill station with reference 18°C and observed mean 19.2°C also has +1.2°C. Their absolute temperatures differ by ten degrees, but their departures from their own references match.

Now use a warmer reference, 28.5°C, for the coastal station. Its anomaly becomes +0.7°C, while the actual July temperature remains 29.2°C. The baseline changed the zero, not the physical observation. If the same constant reference is subtracted throughout a series, differences between its years are unchanged. Combining anomalies based on different reference periods without adjustment can, however, create a misleading comparison. A Celsius anomaly is a temperature difference, not a percentage of the baseline.

Sources: NASA GISS: Temperature anomalies, baselines and data quality ↗ · WMO: Climatological normals ↗

4. Worked case: compare means without hiding variation

Use invented annual temperatures for two five-year periods. Period A has 24, 25, 24, 25 and 24°C, whose mean is 122/5 = 24.4°C. Period B has 25, 24, 26, 25 and 26°C, whose mean is 126/5 = 25.2°C. The later mean is 0.8°C higher. Nevertheless, one later year is cooler than some earlier years. Overlap among individual observations is compatible with a change in their average.

If the period midpoints are ten years apart, the difference corresponds to 0.08°C per year, or 0.8°C per decade, between those block means. This is not a fitted trend through a complete annual record, and two short blocks do not establish a robust regional climate trend. A fitted trend uses the chosen series more fully; its estimate can depend on start and end dates, variability and the statistical method. Plot individual values alongside summaries.

Sources: NASA: How scientists measure global temperature ↗ · NASA GISS: Temperature anomalies, baselines and data quality ↗

5. A physical explanation is more than a rising line

Earth receives energy mainly as sunlight and loses energy as infrared radiation. Greenhouse gases absorb and emit radiation at particular wavelengths. Increasing their concentrations changes the energy balance; the climate system warms until outgoing and incoming energy approach balance under the changed conditions. Human activities, especially fossil-fuel use, have increased greenhouse gases and are the main cause of recent global warming. This explanation combines physics, measured changes and multiple lines of evidence.

Oceans store and redistribute energy, while volcanic aerosols and internal variations such as ENSO can influence shorter-term temperature patterns. Their effects do not require every location or every year to change in the same direction. A local cold spell therefore cannot cancel a global long-term average. Conversely, one hot afternoon cannot by itself quantify human influence on that event; event attribution requires a defined comparison and additional analysis.

Sources: NASA: Physical causes of climate change ↗ · NCERT: World climate and climate change ↗

6. Match the conclusion to the question

An observation describes what was measured. A trend estimates systematic change within a specified record. A weather forecast predicts conditions over a future period from the evolving state of the atmosphere. A climate projection explores possible future statistics under stated assumptions, including emissions scenarios. These products answer different questions. A conditional projection is not a claim that scientists already know every future policy choice or every day’s weather.

For a school data explanation, state the variable, unit, region, years, baseline and source. Show uncertainty and explain whether an apparent jump could come from changing measurements. A moving average can reveal a broad pattern but also hides individual events and does not create new observations. When comparing Indian cities, distinguish local exposure from a national area-weighted estimate. The clearest caption tells readers exactly what the graph can establish.

Sources: WMO: Climatological normals ↗ · NASA: How scientists measure global temperature ↗ · NASA: Physical causes of climate change ↗

Check your understanding