Skip to main content
Bright Minds. Earth Science Earth Science course pack
Resources · The core artifact

The earth science lab notebook.

It is not a worksheet you fill in after the fact. It is the record of the thinking — written at the bench, in pen, with units and significant figures — and it is the one thing in this course no shortcut can fake.

The notebook is the course

In a typical earth science class the lab report is an afterthought — a packet filled out from a worksheet, the answers half-copied from a partner, the conclusion a single sentence written on the bus. In this course the lab notebook is the spine of everything. It is where the prediction is recorded before the measurement, where the observations land in real time, where the calculation is worked out by hand, and where the student finally has to say what the data mean. When the student stands for a lab defense, the notebook is what they defend.

That changes how it must be written. A real earth science notebook is kept in pen, during the investigation, with mistakes struck through by a single line rather than erased — because a crossed-out wrong reading is data too. It is honest, contemporaneous, and complete enough that another earth scientist could repeat the work from it alone. This page lays out exactly what a strong entry contains.

If it isn't written down at the bench, it didn't happen. Memory is not data.

Anatomy of an entry

Every entry in this course follows the same skeleton. Learn it once and it becomes automatic — the structure does the remembering so the student can think about the earth science.

Section What goes here
Title & date A specific title (not "Lab 4") and the calendar date the work was done. One experiment, one dated entry.
Question / purpose One sentence stating what the investigation is meant to find out or measure — e.g. "Identify an unknown mineral from its streak, hardness, luster, and reaction to dilute acid."
Earth Science & prediction The key principle the investigation relies on, plus a specific prediction written before starting — the mineral you expect, the epicenter distance, or the front you expect to track.
Procedure reference A pointer to the written procedure ("see handout, steps 1–7") plus any deviation made on the day. Don't recopy the recipe — record what you actually did differently.
Data tables Measurements as they happen, in ruled tables with a header row naming each quantity, its unit, and the precision of the instrument. Every number gets its units and the right number of significant figures.
Observations Qualitative notes the numbers miss — the moment a streak showed its true color, a specimen fizzed under acid, a cleavage face caught the light. Time-stamped where it matters.
Calculations The math worked by hand with units carried through (dimensional analysis), the key relationship shown, and the final answer rounded to the correct significant figures.
Conclusion A direct answer to the question, compared against the prediction, stated with its uncertainty. Did the result match? If not, why?
Error analysis The real sources of uncertainty — hand-lens judgment, streak-plate wear, an ambiguous hardness scratch — their likely direction and size, and how they would change the result.

And here is that template as a finished entry — one real Experiment Day, kept the way we hold students to. The struck-through note in the margin and the honest sources of error are the point: a real notebook shows the reasoning, not a tidy recopy.

Oct 10 Identifying an unknown mineral
Question
What is mineral sample #4?
Hypothesis
Its metallic look and heft suggest an iron oxide — the streak and hardness tests should decide.
Materials
Sample #4; streak plate; glass plate; steel nail; hand lens; magnet.
Procedure
1. Note luster & color. 2. Streak on unglazed tile. 3. Test hardness against nail & glass. 4. Test with a magnet. ↪ first streak too faint — pressed harder, redid
Observations & data
TestResult
Lustermetallic
Streakdark reddish-brown
Hardness~6 (scratches glass, not nail)
Magnetweakly attracted
Labeled sketch: the reddish-brown streak mark on the plate.
Analysis
Metallic luster + reddish-brown streak + hardness ~6 + weak magnetism point to hematite, not magnetite (which has a black streak and is strongly magnetic).
Conclusion
Sample #4 is hematite — the reddish-brown streak is the deciding test.
Sources of error
The streak was faint at first (too light a pass). Hardness is approximate — “between glass and steel” brackets it near 6.
A model entry. One Experiment Day, kept live at the bench — every section from the template above, in order.

Writing it right: the rules that matter

The structure is half the battle. The other half is a handful of disciplines that separate an earth science notebook from a science-fair poster:

Data tables, sig figs, and error analysis

Three things make an earth science notebook specifically harder — and more valuable — than a general science journal.

Data tables with units and precision. Build the table before the bench work starts, with the columns and units already labeled, so that during a timed mineral ID the student is recording, not designing. Streak color, hardness, luster, cleavage, acid reaction — one row per specimen, each with its unit in the header and its value to the instrument's precision.

Significant figures as a discipline, not a decoration. Sig figs are how an earth scientist tells the truth about precision. Reporting an epicenter distance as "412.7 km" when the seismogram only justifies three figures is a false claim of certainty. The notebook should show the measurement that limits the precision and round the final answer to match it.

Error analysis with direction and size. "Human error" is not error analysis. A real analysis names a specific source — reading a seismogram arrival time to only ±0.5 s, an ambiguous hardness scratch, a streak smeared on a dirty plate — states whether it pushes the result high or low, and estimates how much. This is propagation of uncertainty in plain language, and it is exactly what a lab defense probes.

The lab-notebook defense

At checkpoints the student sits across from the instructor and defends an entry out loud. The questions are simple and devastating to anyone who only copied: Why did you call this mineral quartz and not calcite? What's your prediction based on? Where does the biggest uncertainty come from, and which way does it push your answer? If you ran this again, what would you change? A student who kept the notebook honestly — who wrote the prediction first, recorded in pen, worked the math by hand, and thought about error — answers easily, because the answers are already on the page.

For the criteria the defense is scored against, see the course rubrics. For the safety and readiness routine that makes a strong entry possible in the first place, use the pre-lab checklist before every experiment.

Why this is AI-proof

A language model can write a flawless-sounding lab report. It cannot produce a contemporaneous record of your streak-plate readings, your struck-through misread, the faint fizz you noticed under the acid, or the error analysis that explains why your particular epicenter came in 12 km off. The notebook's value is precisely that it is tied to a real hand at a real bench on a real day — and that the student can defend every line of it from memory. That is not a thing to be outsourced. It is the thing the whole course is built to develop.