Learn

Learning preferences

Make learning more comfortable. Changes apply immediately.

Enlarges page content, including diagrams and their labels. You can also use your browser’s zoom.

Choose the colours that are easiest for you to distinguish. Applies to lesson diagrams and lab plots. Labels, shapes, hatching and line patterns also distinguish meanings. Source photographs and raw image brightness stay unchanged; mask-only views are available for image overlays.

ExampleComparisonSample link

Saved in this browser on this device. Preferences are separate from progress backups.

SCIENCE · INVESTIGATE AND EXPLAIN

Audit recorded weather observations

Compare real station records while preserving timestamps, units, source fields and missing values.

Learning outcome: Separate observation time from retrieval time and reproduce a unit conversion without inventing metadata.

Starting recap

A weather record describes a station at a stated time. A downloaded snapshot preserves that record; retrieval time is when the data were collected for this lesson.

All lessons are open directly. Your saved experiments remain when you leave and return.

Try

KJFK was observed at 19:05 UTC and KSEA at 19:10 UTC on 2026-10-25, so the signed A-minus-B time difference for that order is -5 minutes.

Open the experiment

Predict

Does selecting two records in one panel make their observation times simultaneous?

Experiment

Compare KJFK with KSEA, reverse the pair, and then inspect KDEN. Record timestamps, temperature units and wind-speed conversion, preserving missing fields.

Audit and compare recorded weather observations

Recorded weather observations · frozen source responses · Frozen weather-observations-v1: three deliberately selected NOAA station records dated 25 October 2026, with different observation times and locations. Source values and quality-control codes are retained; missing values stay missing. Calculated humidity is a liquid-water approximation from the reported temperature and dew point, not another observation. No live weather request, pressure-gradient calculation, climate trend or forecast is supplied.

Reset keeps your prediction, captures and practice history.

Recorded temperature and dew point · shared fixed scaleCircle: temperature · square: dew point · connecting segment: their difference0510152025A · KJFK · 19:05 UTCB · KSEA · 19:10 UTCTemperature (°C) · a gap is not a relative-humidity percentage
Frozen NOAA source fields and retained missing values
FieldA · KJFKB · KSEA
Station nameNew York, Kennedy International AirportSeattle, Seattle-Tacoma International Airport
Observation time (UTC)2026-10-25T19:05:00+00:002026-10-25T19:10:00+00:00
Retrieved (UTC)2026-10-25T19:58:30.501772+00:002026-10-25T19:58:31.120219+00:00
Reported station elevation7 m137 m
Temperature18 °C · source QC: V11 °C · source QC: V
Dew point2 °C · source QC: V8 °C · source QC: V
Reported relative humidity34.25 % · source QC: V81.75 % · source QC: V
Wind speed14.832 km/h · source QC: V9.252 km/h · source QC: V
Wind direction200 degrees · source QC: V90 degrees · source QC: V
Wind gustNot reported (null) · source QC: ZNot reported (null) · source QC: Z
API barometricPressure103284.85 Pa · source QC: V101490.06 Pa · source QC: V
API seaLevelPressureNot reported (null) · source QC: ZNot reported (null) · source QC: Z
Precipitation, preceding three hoursNot reported (null) · source QC: ZNot reported (null) · source QC: Z
Source weather descriptionClearCloudy
Raw message fieldEmpty in this response; no raw METAR suppliedEmpty in this response; no raw METAR supplied
Specific observation sourceKJFK observation recordKSEA observation record

Source QC codes are copied verbatim, not converted into a new quality grade. Decimal places are not an uncertainty interval. Missing gust or precipitation is not evidence of calm wind or zero rain. The API's barometricPressure field is preserved by name: this lesson does not assume it is true station pressure or interchangeable with sea-level pressure or an altimeter setting. No pressure gradient is inferred.

Calculations from reported values, not additional measurements
QuantityA · KJFKB · KSEA
Temperature minus dew point16.00 °C3.00 °C
Calculated RH = 100 eₛ(Td)/eₛ(T)34.22%81.74%
Reported RH34.25%81.75%
Wind speed converted by dividing km/h by 3.64.120 m/s2.570 m/s
Signed observation-time difference A − B-5 minutes; negative means A was observed earlier

The calculated humidity reuses the liquid-water saturation approximation from the humidity investigation. A mismatch can reflect rounding or a different source calculation; agreement is not independent sensor validation. These three records cannot establish a station's usual climate, a causal effect of elevation, a temporal warming trend or the future weather.

Inspect weather dataset provenance

NOAA National Weather Service API; snapshot weather-observations-v1. Three deliberately selected airport station records; not simultaneous, random or a climate sample. Frozen records; no live fetch in the lesson. Exact responses, field units and retrieval times are bundled locally. No source icons or external images are loaded. A later source correction does not silently change saved selections.

NOAA API documentation and open-data terms; NWS humidity calculation reference. The API is a recent-observation service; a historical record link can expire while this frozen teaching snapshot remains available.

A − B temperature difference: 7.00 °C. Observation time difference A − B: -5 minutes.

The station and time both change in a cross-station comparison. A reported null is not zero, dry weather or calm wind. Reported humidity and the calculation from temperature/dew point may share inputs; their agreement is not independent validation.

Captures preserve record selections in weather-observations-v1 and replay comparisons of those frozen NOAA reports. Calculations are derived from reported values, not new measurements or live weather updates.

Understand

Start with a traceable record

This activity uses real NOAA National Weather Service API observations preserved in weather-observations-v1. The three deliberately selected records are KJFK, New York Kennedy International Airport; KDEN, Denver International Airport; and KSEA, Seattle-Tacoma International Airport. They are not synthetic weather cases, a random sample or a complete station network.

The lesson reads the frozen local snapshot, not a live feed. Each record retains its official query URL, source fields and retrieval timestamp. The NWS provides observations through its public API, whose documentation describes the service as open data. NWS API documentation.

Create a paper ledger with columns for station identifier, observation timestamp, retrieval timestamp, temperature and unit. Copy a value together with its metadata instead of carrying an unlabeled number into a comparison.

Keep observation and retrieval times separate

All three observations are dated 2026-10-25 in UTC. KJFK is timestamped 19:05, KDEN 19:30 and KSEA 19:10. They were retrieved later, around 19:58:30–31 UTC, with separate exact retrieval timestamps preserved for each record. Retrieval time tells you when the snapshot acquired the record, not when the station observed the weather.

With KJFK as A and KSEA as B, the signed observation-time difference A minus B is −5 minutes. Reverse the pair and the difference becomes +5 minutes. Comparing KDEN with KJFK gives +25 minutes. The sign describes the order of observations; it does not indicate a weather delay or a travel time between stations.

The timestamp suffix +00:00 identifies UTC. Comparing local clock times from different zones without converting them can create a false chronology. This lesson keeps one time basis throughout.

Convert units without changing the observation

The temperature and dew-point fields use degrees Celsius. Wind speed uses kilometers per hour. To convert wind speed to meters per second, divide by 3.6, because 1 km/h = 1000 m/3600 s.

KJFK’s reported wind speed is 14.832 km/h, which converts to 4.12 m/s. KDEN and KSEA each report 9.252 km/h, or 2.57 m/s. This conversion changes the representation, not the measured wind. Additional decimal places in a converted value do not establish additional instrument accuracy.

Write one conversion as a chain of units on paper, then reverse it to recover the source value. Separately record the qualityControl code supplied for that field. The browser preserves those raw codes; it does not transform them into a new quality grade or infer that a retained value is exact.

Do not fill gaps with plausible-looking information

The rawMessage field is an empty string in each saved record. The lesson cannot reconstruct a formal raw METAR message merely because the station is an airport. Preserve the missing message rather than inventing a coded report from selected fields.

Other entries explicitly use null, including all three precipitationLast3Hours values. A missing amount is not zero. A unit code beside a null value identifies the intended quantity but does not supply a measurement. Keep the distinction visible in your ledger.

The exact source records are KJFK at 19:05 UTC, KDEN at 19:30 UTC and KSEA at 19:10 UTC. A live service may not retain an old observation indefinitely; the versioned snapshot preserves the inputs used here.

Describe a comparison precisely

The stored temperatures are 18, 13 and 11°C respectively. KJFK minus KSEA is therefore +7°C for these two observations. This is a comparison between specified records, not a simultaneous spatial map, an all-day temperature difference or a statement about typical climate.

Transfer: separate time and unit errors

A hypothetical record observed at 12:00 UTC is retrieved at 12:20 UTC and reports wind speed 18 km/h. What time should label the observation, and what is its speed in m/s?

Check the weather-observations transfer

The observation time is 12:00 UTC; 12:20 UTC is the retrieval time. The wind speed is 18/3.6 = 5 m/s. Neither conversion nor download time changes what the original observation represents.

Apply

Record A is timestamped 19:05 UTC and B 19:10 UTC on the same date. A minus B is:

Public practice checks a specific claim. It does not establish independent mastery.

Identify the snapshot version and exact source records, explain observation versus retrieval time, and distinguish a missing value from a measured zero.

Continue

Completion requires a saved prediction, at least one capture and a reflection. Presence checks do not grade your reasoning.

Skipping retains experiments and awards no mastery.

Save a science backup or import one

Science work is stored locally in this browser, separately from mathematics and AI. It does not sync automatically. Use this science-only backup; this checkout's older all-subject export does not include science.

A legacy file is available only when the complete snapshot fits the 36 MB limit. For larger snapshots, prepare a complete file set below; never discard notebooks to make a smaller file.

Matching imported notebook drafts replace their current drafts. Omitted lessons and known first attempts remain. Export your current work first if you want to retain both drafts.

Complete file sets and course backups

A file set can hold more work than one science file. Keep its manifest and every part together. Each file is limited to 36 MB. This format needs this version of LearnAI or a compatible newer version; older versions can use only the legacy single-file backup above.

The set includes every saved science course and global continuation at this moment. Later edits need a new export. A conflicting tab exports its own visible work, which may differ from saved work in another tab.

0 file(s) selected. File order and filenames do not determine identity. Missing, changed or conflicting parts block the entire restore; no good part is applied by itself.

Science storage: loading. No account or remote computation is required.