Climate Data, Trends & Attribution · Grade 12 Science
Chapter: Pollution, Carbon & Climate EvidenceClimate Data, Trends & Attribution
Learning goals
- I can interpret anomalies, baselines, and interval trends.
- I can distinguish observations, scenarios, and evidence for attribution.
Readings for this unit
Before this unit
Graphs, averages, greenhouse mechanisms, and proportional rates.
Use the detailed parent plan: two core 50–60-minute sessions per focused unit, with a third assigned for selected calculation/modeling topics. First study the core explanation and Parts 1–2 with their figures/checks; next study Parts 3–4, the whole-unit application, worked cases, varied practice and exit check. Attempt reasoning before opening guidance. Extend time when actual understanding requires it; investigations and chapter assessments have separate sessions.
Start with something to explain
What can a temperature graph establish, and what needs more evidence?
Illustrative evidence case. The complete retained NASA annual series shows fluctuations within a longer historical pattern.
Notice and wonder
Record two observations using the stated quantities or relationships. Keep observations separate from possible causes. Write one question that a new measurement could help answer.
What can a temperature graph establish, and what needs more evidence?
Your first model
Label baseline, interval, global scope and source before calculating.
Build the language from the idea
- anomaly
- A measured or estimated value minus a stated baseline.
- attribution
- Evaluation of the causes of an observed change.
- scenario
- A specified set of assumptions about future drivers.
Weather describes short-term conditions; climate describes statistical patterns over longer periods. A cold day does not disprove a long-term warming trend, and one warm day does not establish it. Specify location, variable, interval, and measurement method before interpreting a climate graph.
An anomaly is a difference from a chosen baseline mean, not an absolute temperature. Changing the baseline shifts anomaly values but does not itself change the underlying trend. A station record, global average, and satellite retrieval have different coverage and methods; compare like quantities.
Estimate a trend from multiple observations using a suitable fit or interval comparison. Natural variability can obscure short sections. A rise of.4°C over 20 years is.02°C/year or.2°C/decade in that interval; do not extrapolate indefinitely without a model and stated scenario.
Attribution asks why an observed change occurred. Mechanisms, observations, and model comparisons can distinguish greenhouse forcing, solar variation, aerosols, land-use change, and internal variability. Correlation alone is not sufficient, but multiple independent lines of physically consistent evidence strengthen a causal explanation.
Uncertainty includes measurement limits, sampling, and model assumptions; it is not equivalent to knowing nothing. Projection ranges can depend on both physical response and future emissions scenarios. Distinguish an observed record from an illustrative dataset and from a conditional projection.
Practice calculations use explicitly supplied hypothetical values; the Learn graph and its full local table additionally retain 55 actual historical NASA GISTEMP annual estimates for 1970–2024, relative to 1951–1980. They are a dated saved selection, not a live current-year feed or a fixed claim about today. Keep baseline, coverage and method/source metadata whenever using observations. Projection values and invented classroom series must remain labeled separately.
Build your understanding
Use the detailed parent plan: two core 50–60-minute sessions per focused unit, with a third assigned for selected calculation/modeling topics. First study the core explanation and Parts 1–2 with their figures/checks; next study Parts 3–4, the whole-unit application, worked cases, varied practice and exit check. Attempt reasoning before opening guidance. Extend time when actual understanding requires it; investigations and chapter assessments have separate sessions.
Part 1 · Define anomalies and their baseline
A temperature anomaly subtracts a reference-period mean from a temperature estimate. This plot uses all 55 complete annual NASA GISTEMP v4 global land–ocean values from 1970 through 2024 in the retained source selection, relative to 1951–1980. An anomaly is not the absolute global temperature and does not imply every location shares the same value. Keeping a consistent baseline supports comparison; changing it shifts the vertical offset without changing differences between the same years.
Read the visual model
Identify the labeled structures and units, then trace the arrows or compare values under the caption’s stated conditions. Separate the displayed evidence from a further inference. Write your reasoning before opening the guidance.
If the baseline were 0.20°C warmer, how would every anomaly change?
Visual reasoning — after your attempt
Each anomaly would decrease by 0.20°C; year-to-year differences and a consistently computed trend would remain unchanged.
A temperature anomaly is a difference from a specified reference average, not a negative absolute temperature. A changed baseline shifts anomaly values but does not by itself change a trend slope when applied consistently. Record the baseline, coverage, units, and aggregation method.
An anomaly of−0.2°C means 0.2°C below the chosen reference mean. It does not indicate that the observed temperature was below freezing.
Complete NASA annual values used in this graph
Complete NASA annual values used in this graph
All 55 annual values are retained in the plotted historical selection. NASA estimates can be revised; this is the saved 2026-10-13 download rather than a live latest-year feed.
| Year | Anomaly (°C relative to 1951–1980) |
|---|---|
| 1970 | 0.03 |
| 1971 | -0.08 |
| 1972 | 0.01 |
| 1973 | 0.16 |
| 1974 | -0.07 |
| 1975 | -0.01 |
| 1976 | -0.1 |
| 1977 | 0.18 |
| 1978 | 0.07 |
| 1979 | 0.16 |
| 1980 | 0.26 |
| 1981 | 0.32 |
| 1982 | 0.14 |
| 1983 | 0.31 |
| 1984 | 0.16 |
| 1985 | 0.12 |
| 1986 | 0.18 |
| 1987 | 0.32 |
| 1988 | 0.39 |
| 1989 | 0.27 |
| 1990 | 0.45 |
| 1991 | 0.41 |
| 1992 | 0.22 |
| 1993 | 0.23 |
| 1994 | 0.31 |
| 1995 | 0.45 |
| 1996 | 0.33 |
| 1997 | 0.47 |
| 1998 | 0.61 |
| 1999 | 0.38 |
| 2000 | 0.39 |
| 2001 | 0.53 |
| 2002 | 0.63 |
| 2003 | 0.62 |
| 2004 | 0.53 |
| 2005 | 0.68 |
| 2006 | 0.64 |
| 2007 | 0.66 |
| 2008 | 0.54 |
| 2009 | 0.66 |
| 2010 | 0.72 |
| 2011 | 0.61 |
| 2012 | 0.65 |
| 2013 | 0.67 |
| 2014 | 0.75 |
| 2015 | 0.89 |
| 2016 | 1.01 |
| 2017 | 0.91 |
| 2018 | 0.85 |
| 2019 | 0.98 |
| 2020 | 1.01 |
| 2021 | 0.85 |
| 2022 | 0.89 |
| 2023 | 1.17 |
| 2024 | 1.28 |
Pause and explain
What information is needed to interpret an anomaly value?
Check your reasoning after answering
The reference period or mean, the measured variable, units, coverage, and averaging method must be identified.
Part 2 · Compare trends over appropriate intervals
A trend summarizes change over a specified interval and method; it is not the same as the difference between two selectively chosen years. These decade means use all ten years in each completed 1980s–2010s decade from the same NASA selection. Averaging reduces some short-term fluctuations but does not remove every uncertainty or establish cause. The 2020–2024 partial decade is not shown as a completed decade because its duration differs.
Read the visual model
Identify the labeled structures and units, then trace the arrows or compare values under the caption’s stated conditions. Separate the displayed evidence from a further inference. Write your reasoning before opening the guidance.
Why avoid labeling the 2020–2024 mean as the complete 2020s decade?
Visual reasoning — after your attempt
Only five years are included, so the interval and sample duration would not match the completed ten-year comparisons.
A simple endpoint slope divides anomaly change by elapsed time, but it discards intermediate variation. Regression or other methods can use the whole record and require assumptions and uncertainty analysis. Selecting convenient start and end years can exaggerate or hide a longer pattern.
A short cooling segment can coexist with a longer warming trend. State the interval rather than treating either selected segment as the entire record.
Pause and explain
Why should a trend estimate not rely on only a favorable endpoint pair?
Check your reasoning after answering
Internal variability and endpoint choice can alter the estimate; a justified interval and full record support a more representative analysis.
Part 3 · Assess coverage and independent indicators
A global estimate combines observations with stated spatial coverage and analysis methods. Coverage bias can arise when sampled places differ systematically from missing places; measurement changes and calibration also require assessment. Independent datasets and indicators help test whether an apparent trend is an artifact of one method, although they need not measure exactly the same physical quantity. Do not invent a station-level error bar from the annual global values alone; use the source’s uncertainty methods.
Read the visual model
Identify the labeled structures and units, then trace the arrows or compare values under the caption’s stated conditions. Separate the displayed evidence from a further inference. Write your reasoning before opening the guidance.
Can ocean heat content replace surface temperature one-for-one?
Visual reasoning — after your attempt
No. It measures a different aspect of the energy system and provides complementary evidence with its own coverage and uncertainty.
Station changes, missing coverage, instruments, and processing can affect a dataset. Responsible analysis documents these issues and compares independent indicators or methods. Agreement among related records supports a pattern but does not make them fully independent if they share source measurements.
Ocean heat and surface temperature measure different parts of the energy response. Their comparison needs physical interpretation, not identical numerical trends.
Pause and explain
Why should data provenance be checked?
Check your reasoning after answering
Shared sources, calibration, sampling, and processing affect independence and uncertainty; multiple labels do not always mean multiple independent lines of evidence.
Part 4 · Separate detection, attribution, and projection
The retained NASA series supports examination of historical global temperature estimates. Attribution additionally compares causal mechanisms, forcings and expected patterns, while projection calculates conditional future behavior under scenarios. A line extrapolated from two endpoints is neither a complete attribution study nor a validated future forecast. Use the whole interval, explicit methods and relevant physical evidence; communicate what is observed, calculated and conditional so a reader can judge each claim.
Read the visual model
Identify the labeled structures and units, then trace the arrows or compare values under the caption’s stated conditions. Separate the displayed evidence from a further inference. Write your reasoning before opening the guidance.
Does the plotted series alone prove a particular future year’s anomaly?
Visual reasoning — after your attempt
No. Future outcomes require assumptions about forcing, response and variability, with uncertainty.
Detecting a trend establishes a change in a record under the analysis assumptions. Attribution evaluates causes using physical models and multiple evidence patterns. Projection applies a model to specified future conditions. A line extrapolated from an illustrative series is neither complete attribution nor a validated forecast.
A fitted slope can be useful for practicing units and describing data while still being insufficient to identify the cause.
Pause and explain
Does a correlation between two time series prove one caused the other?
Check your reasoning after answering
No. Mechanism, alternative explanations, time structure, and other evidence must be considered in an attribution analysis.
Apply the whole unit · Analyze an illustrative anomaly series
Illustrative anomalies at years 0,10,20,30 are 0.0,0.3,0.2,0.6°C relative to one baseline. This is an invented dataset for analysis, not an observational climate record.
Your task
Calculate the 30-year endpoint trend and the 10-to 20-year change, re-express the data with a baseline shifted 0.4°C upward, and explain the limits of extrapolation and attribution.
What your response should show
- Define the system, quantities, units, and relevant conditions before selecting a model.
- Show intermediate calculations or a labeled evidence account, including signs or denominators.
- Explain how the supplied evidence supports the conclusion rather than reporting only an answer.
- Identify a meaningful uncertainty, model limit, or measurement needed to test the conclusion.
Response guidance — after your attempt
Endpoint slope=(0.6−0.0)/30=0.020°C/year=0.20°C/decade.
Years 10–20 show−0.1°C over 10 years, a−0.10°C/decade endpoint slope.
With reference temperature 0.4°C higher, anomalies become−0.4,−0.1,−0.2,+0.2°C; the 30-year slope is unchanged.
The short segment and whole-interval slopes differ without contradiction. Baseline changes shift the record vertically but leave compatible differences unchanged. Intermediate points and uncertainty would matter for a fitted trend, while cause and future response require physical evidence and models beyond this four-point invented series.
Scientific references and source-model foundations
Parent / teacher teaching notes
Require baseline and interval labels. Ask students to distinguish an observed change calculation from an attribution claim or future forecast. Ask the learner to explain each symbol, axis, unit and arrow before calculating. Calculated examples, supplied illustrative records and actual historical observations are labeled separately. Selected public OpenSciEd middle-school source figures are foundation examples, not complete high-school editions or an endorsement; all advanced explanations are supplied locally.
The key idea
Interpret climate claims using baselines, timescales, mechanisms, and dated evidence.
Worked examples
Example 1
- Change=.40°C.
- Divide by 20 years then multiply by 10.
Example 2 · Worked application
- 0.8/40=0.020°C/year.
- Multiply by 10 for 0.20°C/decade.
- This endpoint estimate does not use all intermediate observations like regression does.
Data coverage, baseline and uncertainty affect interpretation.
Example 3
- Original anomaly=1.0°C.
- New anomaly=.8°C.
- Underlying temperature is unchanged.
Example 4 · Find and fix the mistake
- Test the claim against the stated conditions: Data coverage, baseline and uncertainty affect interpretation.
- Endpoints can be sensitive to variability; attribution requires additional evidence and mechanisms.
Data coverage, baseline and uncertainty affect interpretation.
Example 5 · Reason from the visual model
- A global estimate combines observations with stated spatial coverage and analysis methods. Coverage bias can arise when sampled places differ systematically from missing places; measurement changes and calibration also require assessment. Independent datasets and indicators help test whether an apparent trend is an artifact of one method, although they need not measure exactly the same physical quantity. Do not invent a station-level error bar from the annual global values alone; use the source’s uncertainty methods.
- No. It measures a different aspect of the energy system and provides complementary evidence with its own coverage and uncertainty.
Original indicator/method comparison; the local annual file does not contain station coverage or uncertainty estimates.
Avoid these mistakes
- A short noisy interval is not interchangeable with a long-term trend.
Check your understanding
What supports attribution beyond two rising curves?
Discuss the answer
A plausible quantified mechanism, independent observations, and comparisons that test competing explanations.
Return to the phenomenon
Return to the initial case. What can a temperature graph establish, and what needs more evidence? Revise your drawing or explanation using two named ideas and one stated limitation.
Compare explanations after your attempt
Anomalies describe departures from a baseline; detection, attribution and projection are separate evidence tasks.
What should we investigate next?
Which comparison would test a remaining uncertainty? State what you would change, keep comparable and measure.
Investigate, calculate and revise
Keep a first model and an evidence-supported revision. Attempt each task before opening guidance; all required explanations and supplied-data routes are available locally.
Activity 1 · Return to the opening case
Use the opening phenomenon, unit definitions and explained Parts 1–4.
Your task
Return to the initial case. What can a temperature graph establish, and what needs more evidence? Revise your drawing or explanation using two named ideas and one stated limitation.
Success criteria
- Use the stated observations and appropriate units.
- Connect a labeled model with the physical/environmental mechanism.
- State assumptions and an evidence limitation.
- Explain a substantive revision or next test.
Hint
Keep object/material identity, material transfer and energy account distinct.
Response guidance — discuss after your attempt
Anomalies describe departures from a baseline; detection, attribution and projection are separate evidence tasks.
Activity 2 · Transfer to a changed case
Cover the worked answer and use the explained models, relationships and notebook.
Your task
Two absolute global temperature estimates are 14.6°C and 15.0°C. Their baseline is 14.0°C, then changed to 14.2°C. Find both anomaly pairs and show whether their difference changes.
Success criteria
- Use the stated measurements or conditions, with correct units.
- Connect the result to the physical or environmental mechanism or evidence relationship.
- Identify the assumption, model limit or further evidence requested.
Hint
Name known quantities and conditions, show intermediate steps and check conservation.
Response guidance — discuss after your attempt
Original anomalies are 0.6 and 1.0°C; new anomalies 0.4 and 0.8°C. The difference remains 0.4°C, so a uniform baseline shift changes offset rather than physical change.
Words to know
- anomaly
- A measured or estimated value minus a stated baseline.
- attribution
- Evaluation of the causes of an observed change.
- scenario
- A specified set of assumptions about future drivers.
In the real world
Use the chapter investigation to connect this unit to measured evidence and a practical decision.
Exit check
What supports attribution beyond two rising curves?
Check your explanation after answering
A plausible quantified mechanism, independent observations, and comparisons that test competing explanations.
Explain your reasoning without notes. If a key idea is missing, revisit the lesson and try again.