Wait, what? A phone alert says an earthquake was magnitude 6.1. Twenty minutes later, the same event is listed as magnitude 5.9. A few hours after that, the official catalogue says 5.8. Did the earthquake somehow become smaller after it had already happened?
No. The event occurred once in the past. What changed was the scientific estimate describing that event. This Reality Lab teaches a durable evidence habit: when a scientific value is revised, separate the thing in the world from the measurement-and-analysis record that scientists are still improving.
This is exactly the kind of unfamiliar information PSLE Science learners should be able to handle without guessing. The current assessment frame includes interpreting and analysing information, evaluating observations, information and methods, and communicating explanations and reasoning. The 2023 Primary Science syllabus also encourages healthy scepticism: question observations, methods, processes and data, and be willing to update an idea when evidence changes.
Quick Answer
The earthquake did not change after it happened. The estimate changed. Earthquake agencies can publish rapid preliminary magnitude and location estimates using the seismic data available at that moment. As more station data arrive and more detailed analysis is completed, the estimate can be refined. A later value can therefore replace an earlier one without implying that the past physical event grew or shrank after the fact.
The Real-World Evidence Object: One Event, Several Versions
Imagine this original composite earthquake feed:
| Update time | Reported magnitude | Status | Data note |
|---|---|---|---|
| 08:07 | 6.1 | automatic / preliminary | 18 nearby stations available |
| 08:31 | 5.9 | reviewed update | 42 stations available |
| 13:40 | 5.8 | reviewed | additional regional and distant data included |
A learner might look only at the numbers and say, “The earthquake weakened from 6.1 to 5.8.” That sentence confuses a changing record with a changing event. The earthquake’s shaking unfolded in real time and ended. The catalogue entries are later scientific descriptions of that event.
The U.S. Geological Survey explains that magnitudes are often updated in the hours and sometimes days after an earthquake because more data become available and more time-intensive analysis is performed. For large distant events, an initial estimate may be issued before seismic waves have even reached all the stations that will later contribute useful data. That is not a failure of science. It is what timely measurement looks like when the evidence arrives over time.
Owned Learner Job — and the Boundary
The owned job is version-aware evidence reading: decide whether a changing published number means the physical system changed, the measurement changed, the available evidence changed, the analysis method changed, or some combination of these.
This article does not become the main owner of earthquake mechanics, plate tectonics, logarithmic magnitude scales, seismic-wave mathematics, emergency response or hazard advice. It applies existing PSLE inquiry skills to a public scientific communication object: a live event feed whose estimate evolves as evidence improves.
The First Separation: Event Time vs Knowledge Time
There are two timelines hiding inside many science updates.
- Event time: when the physical event happened.
- Knowledge time: when people obtained, processed, reviewed and published information about it.
For an earthquake that began at 08:02, event time is fixed in the past. But knowledge time keeps moving: 08:07 preliminary estimate, 08:31 reviewed estimate, 13:40 refined estimate, perhaps a later catalogue revision.
This distinction appears far beyond seismology. Weather observations are quality-controlled after collection. Satellite datasets are reprocessed. Laboratory values can be reviewed. A species record can be re-identified. A model estimate can be updated as new observations arrive. The learner habit is the same: do not assume every change in a scientific record describes a new physical change in the original object.
Observed, Reported, Inferred
| Layer | What belongs here? | Earthquake example |
|---|---|---|
| Observed | Instrument records | Seismic waveforms recorded at monitoring stations |
| Reported | Processed event parameters | Magnitude, location, depth, origin time and status in an event feed |
| Inferred | Claims made from the reported parameters | Comparisons of event size, likely shaking or significance |
A magnitude update happens in the middle layer: scientists revise an event parameter based on the evidence and analysis. The underlying waveform records may also grow as more stations submit data. The physical earthquake in the first place is not being replayed.
Why Can the First Estimate Be Different?
Rapid public information and final scientific analysis do not always have the same job. A rapid system tries to produce a useful estimate quickly. A later review can use more complete information and more processing time.
The USGS notes several reasons updates can occur:
- not all useful seismic data are available immediately;
- waves take time to travel to more distant stations;
- some contributing data arrive with delays;
- later analysis can use more sophisticated or time-intensive processing;
- human review may replace an automatic preliminary solution;
- archival catalogue work can incorporate still more information.
That creates an important evidence principle: early does not automatically mean wrong, and later does not mean magical certainty. The early estimate may be the best supported value available at that moment. The later estimate can be better supported because its evidence base is broader.
Original Worked Case 1: The “Shrinking Earthquake” Headline
A fictional headline says: “Earthquake Downgraded From 6.2 to 5.8.” A learner comments, “That is strange. The earthquake must have lost energy after it ended.”
The useful word in the headline is downgraded, but even that can be misunderstood. It means the reported estimate was revised downward. It does not mean that the already-completed physical event travelled backward through time and released less energy.
A stronger sentence is: “The agency’s estimate of the earthquake magnitude decreased after additional data and analysis were incorporated.” That sentence identifies the thing that changed: the estimate.
Original Worked Case 2: Two Websites Show Different Values
At 09:00, Website A shows magnitude 5.9 and Website B shows magnitude 6.0 for what appears to be the same event. A student says, “One website must be lying.”
That conclusion is too strong. Before deciding there is a contradiction, check:
- whether both entries refer to the same event;
- the update timestamp;
- whether one site copied an earlier feed;
- which agency produced each solution;
- the magnitude type used;
- whether one solution is automatic and the other reviewed;
- whether both have since converged on a later value.
Scientific disagreement can be real, but a version mismatch is not automatically scientific disagreement. Sometimes the two pages simply froze the evidence at different knowledge times.
Original Worked Case 3: Location Changes Too
A map first places an earthquake under the sea, then a reviewed solution moves the estimated epicentre several kilometres. Did the earthquake physically travel sideways after occurring?
No. The location estimate changed. More station observations and refined analysis can change the estimated source parameters. The same evidence habit applies to magnitude, location and depth: keep the event separate from our evolving estimate of the event.
The Provenance Card Every Learner Should Build
Whenever a live scientific number changes, make a small provenance card:
| Field | Question |
|---|---|
| Object | Which exact event or sample is this record about? |
| Source | Which scientific organisation produced the value? |
| Version time | When was this particular value published or updated? |
| Status | Automatic, preliminary, reviewed, final, provisional or another stated label? |
| Evidence base | What observations were available for this version? |
| Method | Which measurement or analysis method produced the value? |
| Replacement | Has a later official version superseded this one? |
This card prevents a common failure: comparing two numbers without first checking whether they are versions of the same scientific record.
What Evidence Strengthens a Revised Magnitude?
A later estimate is better supported when it uses more relevant station data, improved waveform coverage, appropriate magnitude methods, quality control and expert review. An official event page that documents update time and solution status is much stronger evidence than a screenshot with no timestamp or a copied social-media post.
But “later” is not enough by itself. A random repost made a day later is not stronger merely because its clock time is newer. Provenance matters: which source, which data, which method, which version?
What Would Weaken a Revision Claim?
- No event identifier, location or time is given, so it is unclear whether the posts describe the same earthquake.
- A screenshot removes the update timestamp.
- A news article quotes an early value after the authoritative event page has been revised.
- Two different magnitude types are compared as if they were necessarily identical quantities.
- The later number has no source and may simply be copied from another outlet.
- A learner assumes every numerical difference is measurement error without checking version history.
Alternative Explanations for a Changed Number
When one scientific number becomes another, ask which of these broad explanations fits the evidence:
- The physical system changed. Example: river level genuinely rose between two measurement times.
- The estimate improved. Example: an earthquake magnitude was refined as more seismic data arrived.
- The method changed. Example: a reanalysis used a different algorithm.
- The reference or definition changed. Example: a value was recalculated on a new baseline.
- An error was corrected. Example: a transcription mistake was fixed.
- The records describe different objects. Example: two nearby earthquakes were accidentally compared.
This list is more powerful than memorising “earthquake magnitudes can change.” It transfers to almost any live scientific dashboard.
How Far Can the Conclusion Travel?
If the official magnitude changed from 6.1 to 5.8, you may conclude that the agency revised its magnitude estimate. If the agency explains that more data and analysis were incorporated, you may state that the later estimate is based on a broader or more refined analysis.
You may not conclude from that revision alone that:
- the earthquake physically shrank after it ended;
- the first scientists were careless;
- all early earthquake information is unreliable;
- the later value is exact with zero uncertainty;
- every website still showing the early value is deliberately misleading;
- a smaller revised magnitude means every place experienced weak shaking.
The last point is especially important. Earthquake magnitude is not the same as local shaking intensity. Existing Reality Lab articles own that distinction. This page owns the revision problem.
Tempting but Invalid Reasoning
| Tempting reasoning | What is wrong? | Repair |
|---|---|---|
| “The magnitude fell, so the earthquake became smaller later.” | It confuses event time with knowledge time. | Name the changing object: the estimate. |
| “The first value changed, so it was useless.” | A preliminary estimate can still be useful within its stated status. | Judge evidence according to purpose, timeliness and uncertainty. |
| “The newest number online must be best.” | Recency without provenance is weak evidence. | Prefer an authoritative source with update history. |
| “Two values differ, so one source is dishonest.” | They may be different versions or methods. | Check timestamps, status and method before judging. |
| “Reviewed means impossible to change again.” | Scientific catalogues can still be refined. | Treat status labels as evidence about process, not a promise of eternal immutability. |
PSLE-Style Transfer Case
This is an original practice situation, not a copyrighted examination question.
A monitoring agency posts the following information about Event X:
| Version | Magnitude | Stations used | Status |
|---|---|---|---|
| 1 | 5.7 | 12 | automatic |
| 2 | 5.5 | 31 | reviewed |
A student says, “The earthquake lost magnitude because more stations measured it.”
Evaluate the statement.
A strong answer: “The earthquake did not lose magnitude after it occurred. The reported estimate changed from 5.7 to 5.5 when more station data were included and the event was reviewed. The additional measurements helped refine the estimate of the same past event.”
Why this answer works: it identifies what changed, uses the evidence table, and avoids inventing a mechanism that is not in the data.
Explained Practice
1. Same event or two events?
Two alerts are three minutes apart and have nearby but not identical locations. Before calling the second one a revised magnitude, what must you check?
Answer: Check whether they share the same event identifier or clearly refer to the same earthquake. Otherwise you may be comparing two separate events.
2. Earlier but authoritative
An official agency posted 6.0 at 10:00. A social-media account posted 5.7 at 10:20 with no source. Which is stronger evidence for the event magnitude?
Answer: The official 6.0 is stronger unless the later post can be traced to a newer authoritative revision. Later clock time alone does not create stronger provenance.
3. Preliminary does not mean false
Why is it wrong to translate “preliminary” as “false”?
Answer: Preliminary means the value is an early estimate subject to review or revision. It can still be the best evidence available at that time.
4. What would strengthen confidence?
Name two features that would strengthen confidence in the later solution.
Answer: Examples include more relevant station data, human review, documented method, consistent waveform analysis and an official source with version history.
5. Can a final catalogue ever change?
Why should “final” be read carefully in science?
Answer: “Final” usually refers to a completed stage of the organisation’s workflow, not a claim that future evidence or improved methods can never lead to reanalysis.
Delayed Independent Return
Tomorrow, without reopening this article, answer these:
- What is the difference between event time and knowledge time?
- Why can an earthquake magnitude be revised?
- What does “preliminary” tell you?
- Name one reason two websites can show different values without either one fabricating data.
- Write a scientifically careful sentence describing a revision from 6.1 to 5.8.
Routes to Existing PSLE Science Owners
- How to Tell Observation, Inference, Prediction and Explanation Apart in PSLE Science
- How to Reconcile Two Pieces of PSLE Science Evidence That Seem to Disagree
- How to Read “Most Likely” and “Best Supported” in PSLE Science Without Turning Evidence Into Certainty
- How to Know When PSLE Science Does Not Give Enough Information to Decide
- How Far Can a PSLE Science Conclusion Travel Beyond the Things That Were Actually Tested?
Parent and Tutor Teaching Guide
Teach this with two index cards. On the first, write “What happened?” On the second, write “What do we currently know about what happened?” Ask the learner which card an earthquake magnitude update belongs to. The earthquake belongs to the first card; the changing estimate belongs to the second.
Then transfer the structure. Use a weather forecast that updates, a laboratory result that is reviewed, or a satellite dataset that is reprocessed. Ask whether the changing number describes a changing physical object or a changing knowledge state. This prevents the lesson from becoming earthquake trivia.
For stronger learners, add the provenance card: source, timestamp, status, evidence base, method, replacement. For learners who need a simpler anchor, keep one sentence: “Name what changed.” If the answer is “the estimate”, they have already avoided the central error.
Authoritative Sources and Further Reading
- Singapore Examinations and Assessment Board — 2026 PSLE Science syllabus
- Ministry of Education, Singapore — 2023 Primary Science Teaching and Learning Syllabus
- U.S. Geological Survey — Why/When does the USGS update the magnitude of an earthquake?
- U.S. Geological Survey — Comprehensive Earthquake Catalog (ComCat)
The scientific explanation of earthquake revisions is grounded in these authoritative sources. The fictional feeds, tables and student practice situations above are original teaching examples.
The Quiet Habit to Keep
Science often becomes more accurate by allowing its records to change when evidence improves. That is not weakness. It is one of the ways scientific knowledge stays answerable to reality.
When a number changes, do not ask only, “Which number is right?” Ask first: What changed in the world, what changed in the evidence, and what changed in our estimate? The answer to that question is often the difference between merely reading a science update and actually understanding it.
