How do you translate volcanic ash advisories correctly? Start by treating every Volcanic Ash Advisory (VAA) as a time-stamped, geographically bounded aviation-information product, not as ordinary weather prose. A reliable translation must preserve the issuing Volcanic Ash Advisory Centre (VAAC), volcano identity, UTC times, observed ash cloud, forecast positions, flight levels, uncertainty, and next-advisory instruction. Translating VAAC forecasts, aviation warnings and flight-level abbreviations accurately is a safety-sensitive technical communication task.
People searching for how to translate volcanic ash warnings, what VAA means, how to read VAAC reports, volcanic ash forecast translation or how to interpret flight levels often stumble over the same hidden problem: the abbreviation looks small, but its role is large. An ash cloud reported as not identifiable from satellite data is not proof that no ash exists. A forecast for six hours after an observation is not a current observation. A flight level is a pressure-altitude reference, not simply a terrain elevation.
This guide teaches a first-principles, repeatable translation method for students, translators, aviation communicators and international readers. It explains the difference between advisory text and graphical ash forecasts, uses carefully labelled fictional examples rather than live flight instructions, tests the common failure modes, and shows how to hand off uncertainty without pretending to make operational decisions. For real aviation operations, the current official products and approved procedures always control.
The one-minute answer
A correct volcanic ash advisory translation preserves the product’s identity, time, geography, vertical extent, forecast horizon and information status. Translate explanatory language into natural target-language prose, but keep established codes, coordinates, identifiers and quantitative values exact unless an authorised display convention requires a documented change. Never add a certainty that the source lacks, never convert forecast ash into observed ash, and never infer a safe route from the translation alone. A translation is an aid to understanding, not an independent flight clearance or replacement for an official briefing.
The visible translation problem
Imagine a dispatcher reads a note that says the ash cloud is not identifiable from satellite imagery, then sees a translation announcing that the ash has cleared. The translated sentence appears reassuring, but it has converted limited observation into a negative finding. In another example, the source says forecast ash at plus twelve hours and the target text says current ash. That change can move an event along the timeline by half a day. A third case changes FL250 into 2,500 feet. The numbers look related, yet the vertical reference has been ruined. The problem is not elegant phrasing; it is preservation of operational evidence.
What this article owns—and what it does not
This article owns VAA and VAG text, VAAC observations, forecast lead times, volcanic ash cloud polygons and flight-level wording. It does not replace the separate eduKateSG article about NOTAM runway and flight restrictions, the METAR/TAF weather-report guide or general emergency-alert translation. Those are complementary documents with different issuing authorities and different canonical jobs.
First principles: how an aviation advisory preserves meaning
1. An advisory is not a flight clearance
A VAA provides technical information about volcanic ash; it is not an instruction that every aircraft must take a particular route. The translation must not silently promote advisory information into a mandatory operational command.
A good translation makes this principle testable. Write down the exact source field that supports the claim, then compare the translated statement against the issuing document. Ask what a reader could mistakenly infer if the source’s qualifier disappeared. In this context, ‘an advisory is not a flight clearance’ is not a matter of style: it governs the difference between reporting a published aviation-information product and constructing a new assertion that nobody issued.
2. The source and issue time belong together
A product’s date-time group identifies when the issuing office released information. Keep this separate from observed-cloud time, forecast-valid time and next-issue time, because readers make different decisions from each.
A good translation makes this principle testable. Write down the exact source field that supports the claim, then compare the translated statement against the issuing document. Ask what a reader could mistakenly infer if the source’s qualifier disappeared. In this context, ‘the source and issue time belong together’ is not a matter of style: it governs the difference between reporting a published aviation-information product and constructing a new assertion that nobody issued.
3. Observed and forecast are separate evidence states
An observed ash cloud describes information for an observation time. Forecast ash concerns a later target time and can change with new analysis. Preserve the labels even when the target language prefers a different sentence order.
A good translation makes this principle testable. Write down the exact source field that supports the claim, then compare the translated statement against the issuing document. Ask what a reader could mistakenly infer if the source’s qualifier disappeared. In this context, ‘observed and forecast are separate evidence states’ is not a matter of style: it governs the difference between reporting a published aviation-information product and constructing a new assertion that nobody issued.
4. Vertical extent has a reference system
FL values conventionally refer to standard-pressure flight levels, not metres above a volcano or height above the nearest airport. Decode only in an explanatory note when needed and preserve the original FL token.
A good translation makes this principle testable. Write down the exact source field that supports the claim, then compare the translated statement against the issuing document. Ask what a reader could mistakenly infer if the source’s qualifier disappeared. In this context, ‘vertical extent has a reference system’ is not a matter of style: it governs the difference between reporting a published aviation-information product and constructing a new assertion that nobody issued.
5. Coordinates require geometry, not impression
A polygon made from several latitude and longitude points is a spatial outline, not a place name. The point order, hemispheres and numeric precision can materially change the represented airspace.
A good translation makes this principle testable. Write down the exact source field that supports the claim, then compare the translated statement against the issuing document. Ask what a reader could mistakenly infer if the source’s qualifier disappeared. In this context, ‘coordinates require geometry, not impression’ is not a matter of style: it governs the difference between reporting a published aviation-information product and constructing a new assertion that nobody issued.
6. Uncertainty is a finding, not a failure
Not identifiable, not available, not provided, unknown and none are not interchangeable. Each one describes a different knowledge condition and must remain visibly different in translation.
A good translation makes this principle testable. Write down the exact source field that supports the claim, then compare the translated statement against the issuing document. Ask what a reader could mistakenly infer if the source’s qualifier disappeared. In this context, ‘uncertainty is a finding, not a failure’ is not a matter of style: it governs the difference between reporting a published aviation-information product and constructing a new assertion that nobody issued.
7. Product family and authority matter
VAAC advisory, volcanic ash graphic, volcanic ash SIGMET, volcano observatory notice and NOTAM are separate product types. A translator must preserve who is speaking and which document the audience is reading.
A good translation makes this principle testable. Write down the exact source field that supports the claim, then compare the translated statement against the issuing document. Ask what a reader could mistakenly infer if the source’s qualifier disappeared. In this context, ‘product family and authority matter’ is not a matter of style: it governs the difference between reporting a published aviation-information product and constructing a new assertion that nobody issued.
8. Updates are part of the meaning
An advisory number, next advisory field and no-further-advisories statement govern document continuity. A newer product may revise an earlier forecast, so translation should never present archived text as today’s active situation.
A good translation makes this principle testable. Write down the exact source field that supports the claim, then compare the translated statement against the issuing document. Ask what a reader could mistakenly infer if the source’s qualifier disappeared. In this context, ‘updates are part of the meaning’ is not a matter of style: it governs the difference between reporting a published aviation-information product and constructing a new assertion that nobody issued.
A repeatable ten-step translation workflow
Step 1: Identify the exact product
Record whether the source is VAA text, a VAG graphical depiction or another related aviation product. Copy the issuing VAAC, advisory reference and original source URL into your job sheet.
The exit test for this step is practical: another reader with the same source should be able to reproduce your decision without guessing which value, source or time you meant. Place uncertainties in a review log rather than quietly repairing them. When the original does not contain enough information, note the gap and move to authorised clarification. This habit protects both the accuracy of the translation and the separation between language work and operational judgement.
Step 2: Capture the source timestamp
Copy the date-time group verbatim, identify UTC or Z, and note separately any publication-page timestamp. Do not adjust the clock without a clearly stated and independently checked conversion.
The exit test for this step is practical: another reader with the same source should be able to reproduce your decision without guessing which value, source or time you meant. Place uncertainties in a review log rather than quietly repairing them. When the original does not contain enough information, note the gap and move to authorised clarification. This habit protects both the accuracy of the translation and the separation between language work and operational judgement.
Step 3: Separate observation from forecast
Mark each line OBS or FCST, including offsets such as +6HR, +12HR and +18HR. If the original gives no observation, do not invent one to make the narrative read smoothly.
The exit test for this step is practical: another reader with the same source should be able to reproduce your decision without guessing which value, source or time you meant. Place uncertainties in a review log rather than quietly repairing them. When the original does not contain enough information, note the gap and move to authorised clarification. This habit protects both the accuracy of the translation and the separation between language work and operational judgement.
Step 4: Build the place-and-altitude map
Extract volcano location, point coordinates, flight-level bounds, direction of movement and ash-cloud geometry. Keep each measurement beside its label in a structured ledger.
The exit test for this step is practical: another reader with the same source should be able to reproduce your decision without guessing which value, source or time you meant. Place uncertainties in a review log rather than quietly repairing them. When the original does not contain enough information, note the gap and move to authorised clarification. This habit protects both the accuracy of the translation and the separation between language work and operational judgement.
Step 5: Define abbreviations before rewriting
Check VAAC, VAA, VAG, FL, OBS, FCST, PSN, DTG and RMK against authoritative document context. Expand in explanatory prose only when doing so helps the intended reader.
The exit test for this step is practical: another reader with the same source should be able to reproduce your decision without guessing which value, source or time you meant. Place uncertainties in a review log rather than quietly repairing them. When the original does not contain enough information, note the gap and move to authorised clarification. This habit protects both the accuracy of the translation and the separation between language work and operational judgement.
Step 6: Draft the target-language message
Use short clauses that preserve speaker, timestamp, evidence state, affected area and explicit uncertainty. Keep codes and values in their original fields where practical.
The exit test for this step is practical: another reader with the same source should be able to reproduce your decision without guessing which value, source or time you meant. Place uncertainties in a review log rather than quietly repairing them. When the original does not contain enough information, note the gap and move to authorised clarification. This habit protects both the accuracy of the translation and the separation between language work and operational judgement.
Step 7: Run a numeric and semantic comparison
Compare the target against source for every identifier, time, coordinate, forecast horizon and negative expression. This is a bilingual audit, not a general spellcheck.
The exit test for this step is practical: another reader with the same source should be able to reproduce your decision without guessing which value, source or time you meant. Place uncertainties in a review log rather than quietly repairing them. When the original does not contain enough information, note the gap and move to authorised clarification. This habit protects both the accuracy of the translation and the separation between language work and operational judgement.
Step 8: Check the product boundary
Make sure the translated text has not become a forecast issued by the translator, a safety assurance, a closure order or a route recommendation absent from the official source.
The exit test for this step is practical: another reader with the same source should be able to reproduce your decision without guessing which value, source or time you meant. Place uncertainties in a review log rather than quietly repairing them. When the original does not contain enough information, note the gap and move to authorised clarification. This habit protects both the accuracy of the translation and the separation between language work and operational judgement.
Step 9: Apply independent specialist review
In a live aviation context use a trained reviewer familiar with the official product family, the source language and the receiving organisation’s approved procedures.
The exit test for this step is practical: another reader with the same source should be able to reproduce your decision without guessing which value, source or time you meant. Place uncertainties in a review log rather than quietly repairing them. When the original does not contain enough information, note the gap and move to authorised clarification. This habit protects both the accuracy of the translation and the separation between language work and operational judgement.
Step 10: Release with provenance
Keep issue identifier, source URL, retrieval time, target-language version, reviewer and a note on any uncertainty. Mark educational examples as fictional and any archival source as archival.
The exit test for this step is practical: another reader with the same source should be able to reproduce your decision without guessing which value, source or time you meant. Place uncertainties in a review log rather than quietly repairing them. When the original does not contain enough information, note the gap and move to authorised clarification. This habit protects both the accuracy of the translation and the separation between language work and operational judgement.
Field dictionary: translate these labels without changing their data
VAAC name
Preserve the issuing centre’s proper name and distinguish it from adjacent agencies. Check whether another centre is being quoted rather than acting as primary issuer.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original VAAC name entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
Volcano name and number
Keep both proper name and official identifier when present. Do not silently substitute a nearby volcano with a similar spelling.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original Volcano name and number entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
PSN
Preserve the volcano’s published position and coordinate precision. A missing N or S reverses the hemisphere.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original PSN entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
DTG
Preserve the advisory date-time group and UTC convention. Publication time is not necessarily observation time.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original DTG entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
ADVISORY NR
Retain year and sequence identity, including leading zeros. A sequence is an identifier, not a decimal quantity.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original ADVISORY NR entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
INFO SOURCE
Preserve the named evidence source, such as satellite analysis or another VAAC. Source attribution is not a declaration that the translator observed the event.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original INFO SOURCE entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
ERUPTION DETAILS
State only the eruption details actually supplied. Not provided is not the same as no eruption.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original ERUPTION DETAILS entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
OBS VA DTG
Keep the observed ash time separate from document issue time. A target translation can rearrange the sentence but not the time role.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original OBS VA DTG entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
OBS VA CLD
Keep any explicit observation statement, cloud geometry and altitude. Not identifiable from satellite imagery is not a claim of clear air.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original OBS VA CLD entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
FCST VA CLD +6HR
Mark this as a forecast for a specified lead time. Do not call it a verified six-hour observation.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original FCST VA CLD +6HR entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
FCST VA CLD +12HR
Keep the twelve-hour horizon and any polygon description. Forecast lead time must not become a clock time without calculation and attribution.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original FCST VA CLD +12HR entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
FCST VA CLD +18HR
Keep the eighteen-hour horizon and source conditionality. An omitted forecast should stay omitted or labelled unavailable.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original FCST VA CLD +18HR entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
FL notation
Retain flight levels and any top/bottom markers. FL350 is not 3,500 feet and is not volcano summit height.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original FL notation entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
RMK and NXT ADVISORY
Preserve remarks and announced update or closure instructions. No further advisories is not itself proof of zero airborne ash.
Keep the source field beside its target-language explanation in a two-column audit sheet. A reviewer should be able to point to the original RMK and NXT ADVISORY entry, see exactly how it appears in the translated version, and confirm that no unstated assumption entered the result. This is especially important when a general reader may understand the English sentence but not the conventional aviation notation.
Worked examples: twenty-eight specific translation decisions
The following scenarios are invented for teaching. They are not live aviation products, not real flight instructions and not substitutes for an official source. Their purpose is to show how a translator preserves facts while building clear natural-language explanations.
Worked example 1: The date-time group
Source fragment, given here as an illustrative English gloss: “DTG: 20261026/1155Z”. A faithful target-language explanatory rendering is: “Advisory dated 26 October 2026 at 11:55 UTC”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is confusing advisory issue time with time of the volcanic eruption. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Check the digits, year, month, day and Z notation against the original header. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 2: A named centre
Source fragment, given here as an illustrative English gloss: “VAAC: ANCHORAGE”. A faithful target-language explanatory rendering is: “Issued by Anchorage Volcanic Ash Advisory Center”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is replacing the originator with another meteorological office mentioned in remarks. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Confirm the header and centre responsibility field. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 3: A volcano identifier
Source fragment, given here as an illustrative English gloss: “VOLCANO: EXAMPLE PEAK 000123”. A faithful target-language explanatory rendering is: “Volcano named Example Peak, official example identifier 000123”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is treating 000123 as an eruption intensity or removing leading zeros. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Read it as a code and retain the fictional label. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 4: Northern longitude and latitude
Source fragment, given here as an illustrative English gloss: “PSN: N5019 E155027”. A faithful target-language explanatory rendering is: “Position given by the source as N5019 E155027; decode only using the issuing product’s coordinate convention”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is swapping east and west or freely rounding coordinate digits. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Verify original coordinate parsing with the issuing product guidance. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 5: Summit elevation
Source fragment, given here as an illustrative English gloss: “SOURCE ELEV: 5842 FT AMSL”. A faithful target-language explanatory rendering is: “Source elevation 5,842 feet above mean sea level”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is confusing mountain height with ash cloud flight level. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Check the named altitude reference and original units. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 6: Unknown eruption details
Source fragment, given here as an illustrative English gloss: “ERUPTION DETAILS: NOT PROVIDED”. A faithful target-language explanatory rendering is: “The advisory does not provide eruption details”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is changing missing information into a statement that no eruption occurred. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Preserve the absence of details rather than a negative claim. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 7: Unseen ash on imagery
Source fragment, given here as an illustrative English gloss: “OBS VA CLD: VA NOT IDENTIFIABLE FROM SATELLITE”. A faithful target-language explanatory rendering is: “Volcanic ash is not identifiable from the available satellite analysis”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is turning non-identification into confirmation of no volcanic ash. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Check for observational limitations and any other listed evidence. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 8: Missing observation time
Source fragment, given here as an illustrative English gloss: “OBS VA DTG: NOT PROVIDED”. A faithful target-language explanatory rendering is: “An observation date-time is not provided”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is inventing an observation time from the issue header. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Audit the observation field independently of DTG. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 9: Forecast lead time six
Source fragment, given here as an illustrative English gloss: “FCST VA CLD +6HR: FL200/FL320”. A faithful target-language explanatory rendering is: “Six-hour forecast mentions an ash cloud between flight levels FL200 and FL320”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is converting a forecast into a current cloud measurement. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Verify product timing and the paired lower and upper level. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 10: Forecast lead time twelve
Source fragment, given here as an illustrative English gloss: “FCST VA CLD +12HR: MOV E”. A faithful target-language explanatory rendering is: “Twelve-hour forecast indicates movement east”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is reporting that ash was observed travelling east at issue time. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Confirm forecast label and movement direction. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 11: Forecast lead time eighteen
Source fragment, given here as an illustrative English gloss: “FCST VA CLD +18HR: NOT PROVIDED”. A faithful target-language explanatory rendering is: “No eighteen-hour ash forecast is provided in this example”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is treating an absent forecast as a zero-risk forecast. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Preserve not provided as an explicit data gap. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 12: Forecast polygon
Source fragment, given here as an illustrative English gloss: “FCST VA CLD: N4700 E15000 – N4630 E15130 – N4600 E15045”. A faithful target-language explanatory rendering is: “A forecast ash area described by an ordered set of positions”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is reducing the polygon to a single city or omitting one corner. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Compare every latitude and longitude point in order. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 13: Wind movement
Source fragment, given here as an illustrative English gloss: “MOV NE 15KT”. A faithful target-language explanatory rendering is: “Motion to the northeast at a reported speed of 15 knots”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is calling movement toward the southwest or changing knots into kilometres per hour without conversion. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Check compass direction and speed unit. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 14: Cloud top altitude
Source fragment, given here as an illustrative English gloss: “TOP FL380”. A faithful target-language explanatory rendering is: “Top of the reported ash cloud at flight level 380”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is calling this 380 feet or a precise height above the volcano. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Keep FL380 as an operational notation. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 15: Lower level
Source fragment, given here as an illustrative English gloss: “SFC/FL140”. A faithful target-language explanatory rendering is: “A vertical range expressed from surface to flight level 140”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is substituting the volcano’s summit level for the surface boundary. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Keep the two different boundaries visible. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 16: Remarks about evidence
Source fragment, given here as an illustrative English gloss: “RMK: SATELLITE DATA INCONCLUSIVE”. A faithful target-language explanatory rendering is: “Remarks state that satellite analysis is inconclusive”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is calling the ash absent or the eruption confirmed. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Translate the degree of confidence, not a guessed conclusion. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 17: Another VAAC cited
Source fragment, given here as an illustrative English gloss: “INFO SOURCE: VAAC TOKYO”. A faithful target-language explanatory rendering is: “Tokyo VAAC is cited as an information source”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is silently altering the issuing vaac header to tokyo. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Distinguish reported source from issuer. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 18: Next advisory time
Source fragment, given here as an illustrative English gloss: “NXT ADVISORY: 20261026/1800Z”. A faithful target-language explanatory rendering is: “Next advisory expected at 18:00 UTC on 26 October 2026”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is treating the next issuance as a forecast valid time. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Cross-check with NXT ADVISORY field. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 19: No further advisories
Source fragment, given here as an illustrative English gloss: “NXT ADVISORY: NO FURTHER ADVISORIES”. A faithful target-language explanatory rendering is: “The centre is not planning further advisories under this sequence”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is assuring readers that aircraft operations are unrestricted. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Retain the administrative meaning without operational inference. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 20: Estimated location
Source fragment, given here as an illustrative English gloss: “PSN: ESTIMATED”. A faithful target-language explanatory rendering is: “Location is an estimate, not a confirmed exact point”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is discarding the qualifier because the coordinate looks precise. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Carry estimated language beside the value. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 21: Archived advisory
Source fragment, given here as an illustrative English gloss: “ADVISORY NR: 2024/017”. A faithful target-language explanatory rendering is: “An advisory in the 2024 series, number 017”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is labelling the historical record as current 2026 guidance. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: State the historical date and retrieval context. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 22: Ash versus ashfall
Source fragment, given here as an illustrative English gloss: “OBS VA CLD: ASH ALOFT”. A faithful target-language explanatory rendering is: “Volcanic ash described as airborne”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is translating airborne ash as ash deposited on the ground. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Preserve the distinction between atmospheric cloud and ground deposit. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 23: Graphical contours
Source fragment, given here as an illustrative English gloss: “VAG: OBS +6HR +12HR +18HR PANELS”. A faithful target-language explanatory rendering is: “A volcanic ash graphic separates observation and forecast panels”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is treating every image panel as a simultaneous snapshot. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Label panel time and forecast status in alt text or caption. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 24: Partial product message
Source fragment, given here as an illustrative English gloss: “FCST VA CLD +12HR: DATA UNAVAILABLE”. A faithful target-language explanatory rendering is: “The twelve-hour forecast data are unavailable in the example”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is automatically copying the six-hour values forward. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Request the complete current source before filling any gap. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 25: Unit conversion caution
Source fragment, given here as an illustrative English gloss: “TOP FL300; SOURCE ELEV 1800 M”. A faithful target-language explanatory rendering is: “Ash top given as FL300; source elevation reported in metres”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is converting both to the same-looking altitude label. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Do not erase the different reference frames. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 26: Geographic handoff
Source fragment, given here as an illustrative English gloss: “VAAC BOUNDARY HANDOFF IN REMARKS”. A faithful target-language explanatory rendering is: “Remarks describe a transfer of product coordination across centre areas”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is treating a forecast change at a boundary as a new volcano. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Track advisory chronology and issuing-office transition. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 27: Contradictory source feeds
Source fragment, given here as an illustrative English gloss: “SATELLITE NOT IDENTIFIABLE; PILOT REPORT RECEIVED”. A faithful target-language explanatory rendering is: “The advisory records differing evidence channels without reconciling them into certainty”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is deleting one channel to create a simple story. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Preserve source attribution for each observation. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Worked example 28: Versioned translation
Source fragment, given here as an illustrative English gloss: “TRANSLATION OF VAA 2026/004, RETRIEVED 1205Z”. A faithful target-language explanatory rendering is: “A translated copy tied to a specific advisory and retrieval time”. The aim is not to copy the source typography into every language, but to preserve the specific information carried by that field. A translator can reorganise ordinary prose to make it readable, yet the data element must remain bound to its proper authority, unit, time and evidence status.
The tempting error is copying the translation into a new bulletin without indicating staleness. This may arise from fluent rewriting, an automatic translation tool, a reviewer who knows ordinary weather vocabulary but not aviation products, or a design change that separates a label from its value. In each case the surface may look polished while the reader receives materially different information. A responsible translation therefore prioritises semantic equivalence over a cosmetic appearance of completeness.
Verification action: Include provenance and the live official source. Compare the English explanation with the original line character by character for protected tokens and clause by clause for status words. If the source provides a qualifier, carry it across; if a crucial item is absent, mark it absent. The final question is whether a reader could reconstruct the same time, location, evidence and scope from the target wording without being led to a stronger claim.
Transfer challenge: rewrite this example for a public explainer, an expert handoff and a bilingual training worksheet. The public explainer may define the technical expression once; the expert handoff should retain conventional code and sequence; the worksheet should ask learners to distinguish source evidence from inference. In all three settings, keep the same substantive claim. Variation in voice is allowed, variation in the documented measurement or qualification is not.
Failure modes and precise repairs
1. Swapping observation and forecast
Keep OBS and FCST visibly separated wherever the source makes the distinction; an attractive paragraph should never erase the temporal status.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
2. Calling undetected ash absent
A sensor limitation cannot prove a negative. Preserve not identifiable, not detected and not provided with their distinct meanings.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
3. Dropping flight-level context
A bare number cannot stand in for FL. Retain flight-level notation and avoid invented height conversions.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
4. Rounding coordinates casually
A small numerical change can move the polygon. Treat sign, precision, sequence and hemisphere as protected data.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
5. Confusing eruption with plume
The eruption date and ash-cloud observation date may differ. A translation should not collapse these into one timestamp.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
6. Changing issued-by attribution
The agency that supplied information may not be the agency that released the product. Attribute both accurately.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
7. Turning the centre into an authority to fly
A VAAC provides information; the translation should not masquerade as an operational clearance.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
8. Overwriting a missing forecast
Do not infer +18 hours from +12 hours. Report missing and direct readers to official products.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
9. Translating every acronym as ordinary vocabulary
Some short codes are machine- or sector-recognised. Explain them without destroying the original.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
10. Presenting a historical VAA as live
Preserve issue number and date. Public archives are evidence of past products, not current instructions.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
11. Mistaking ashfall for airborne ash
One concerns deposits at the surface, another airborne hazard; preserve altitude and location context.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
12. Stripping update information
The next-advisory field tells readers what continuity to expect, not whether all hazards have ceased.
Repair method: identify the single source field at risk, compare its value and role with the translated sentence, then rewrite only the faulty span. Do not redesign the whole document to conceal a local mismatch. Log the correction so later reviewers can tell whether a wording change was stylistic or corrected a material factual drift.
Guided practice: learn, retrieve, transfer
Use these as simulated source-reading tasks. Try each before looking at the model answer. The goal is to make the error visible, not simply to memorise a list of abbreviations. Score yourself separately for source identification, correct interpretation, numeric fidelity, uncertainty and reader-facing clarity.
Practice 1: The date-time group
Prompt: a bilingual note contains “DTG: 20261026/1155Z”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “Advisory dated 26 October 2026 at 11:55 UTC”. A common wrong inference is confusing advisory issue time with time of the volcanic eruption. To verify the correction, check the digits, year, month, day and Z notation against the original header. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 2: A named centre
Prompt: a bilingual note contains “VAAC: ANCHORAGE”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “Issued by Anchorage Volcanic Ash Advisory Center”. A common wrong inference is replacing the originator with another meteorological office mentioned in remarks. To verify the correction, confirm the header and centre responsibility field. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 3: A volcano identifier
Prompt: a bilingual note contains “VOLCANO: EXAMPLE PEAK 000123”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “Volcano named Example Peak, official example identifier 000123”. A common wrong inference is treating 000123 as an eruption intensity or removing leading zeros. To verify the correction, read it as a code and retain the fictional label. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 4: Northern longitude and latitude
Prompt: a bilingual note contains “PSN: N5019 E155027”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “Position given as 50°19′ north, 155°02′7″ east in the source’s compact format”. A common wrong inference is swapping east and west or freely rounding coordinate digits. To verify the correction, verify original coordinate parsing with the issuing product guidance. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 5: Summit elevation
Prompt: a bilingual note contains “SOURCE ELEV: 5842 FT AMSL”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “Source elevation 5,842 feet above mean sea level”. A common wrong inference is confusing mountain height with ash cloud flight level. To verify the correction, check the named altitude reference and original units. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 6: Unknown eruption details
Prompt: a bilingual note contains “ERUPTION DETAILS: NOT PROVIDED”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “The advisory does not provide eruption details”. A common wrong inference is changing missing information into a statement that no eruption occurred. To verify the correction, preserve the absence of details rather than a negative claim. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 7: Unseen ash on imagery
Prompt: a bilingual note contains “OBS VA CLD: VA NOT IDENTIFIABLE FROM SATELLITE”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “Volcanic ash is not identifiable from the available satellite analysis”. A common wrong inference is turning non-identification into confirmation of no volcanic ash. To verify the correction, check for observational limitations and any other listed evidence. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 8: Missing observation time
Prompt: a bilingual note contains “OBS VA DTG: NOT PROVIDED”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “An observation date-time is not provided”. A common wrong inference is inventing an observation time from the issue header. To verify the correction, audit the observation field independently of DTG. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 9: Forecast lead time six
Prompt: a bilingual note contains “FCST VA CLD +6HR: FL200/FL320”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “Six-hour forecast mentions an ash cloud between flight levels FL200 and FL320”. A common wrong inference is converting a forecast into a current cloud measurement. To verify the correction, verify product timing and the paired lower and upper level. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 10: Forecast lead time twelve
Prompt: a bilingual note contains “FCST VA CLD +12HR: MOV E”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “Twelve-hour forecast indicates movement east”. A common wrong inference is reporting that ash was observed travelling east at issue time. To verify the correction, confirm forecast label and movement direction. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 11: Forecast lead time eighteen
Prompt: a bilingual note contains “FCST VA CLD +18HR: NOT PROVIDED”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “No eighteen-hour ash forecast is provided in this example”. A common wrong inference is treating an absent forecast as a zero-risk forecast. To verify the correction, preserve not provided as an explicit data gap. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 12: Forecast polygon
Prompt: a bilingual note contains “FCST VA CLD: N4700 E15000 – N4630 E15130 – N4600 E15045”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “A forecast ash area described by an ordered set of positions”. A common wrong inference is reducing the polygon to a single city or omitting one corner. To verify the correction, compare every latitude and longitude point in order. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 13: Wind movement
Prompt: a bilingual note contains “MOV NE 15KT”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “Motion to the northeast at a reported speed of 15 knots”. A common wrong inference is calling movement toward the southwest or changing knots into kilometres per hour without conversion. To verify the correction, check compass direction and speed unit. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Practice 14: Cloud top altitude
Prompt: a bilingual note contains “TOP FL380”. Explain this in plain English for a non-specialist without discarding the source code. Then describe one wrong inference a rushed reader might make.
Model answer: “Top of the reported ash cloud at flight level 380”. A common wrong inference is calling this 380 feet or a precise height above the volcano. To verify the correction, keep FL380 as an operational notation. Finally, explain whether the message reports an observation, forecasts a later state or merely describes metadata; if the excerpt alone cannot establish that role, say so.
Frequently asked questions
Can I translate VAAC abbreviations into ordinary words?
You may explain them in running prose, but preserve the exact identifiers and conventional codes within the translated document when they carry operational meaning.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
Does not identifiable mean no volcanic ash?
No. It describes an observational limitation. Another sensor, report or later analysis could identify ash that one product could not.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
Is FL300 simply an altitude of 300 feet?
No. Flight levels are referenced to a standard pressure setting; the three-digit number is not a direct number of feet.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
May I convert UTC to local time?
Only as a clearly labelled supplementary explanation checked against the date and daylight-saving rules; preserve source UTC in the working record.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
What is the difference between VAA and VAG?
A VAA is advisory text and a VAG is a graphical product depicting related ash information. Their formats differ, even when they communicate the same event.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
Is a VAA the same as a SIGMET?
No. A volcanic ash SIGMET is an operational meteorological warning issued through its own process. Do not replace one product name with the other.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
What if a forecast field is not provided?
Keep it missing or explicitly unavailable. Do not fill it from an older product, nearby forecast or assumption.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
Should I translate a volcano name?
Use the official identifier and established proper name. An explanatory local-language form may be added only if it cannot be confused with a different volcano.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
What if the source advisory has a suspected error?
Preserve source evidence in the translation, flag the discrepancy to the issuer or responsible specialist, and avoid an unapproved silent correction.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
Is the most recent web search result always the latest operational bulletin?
No. Archive pages, caches and product displays can lag. Use the official issuing service and its timestamps.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
How should classroom exercises handle safety risk?
Mark all simulated bulletins as fictional and prevent their reuse as live instructions or aviation advisories.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
Who should review a high-stakes translation?
A competent bilingual technical reviewer and the organisation’s authorised aviation specialist should verify the product before operational distribution.
The rule of thumb is to preserve what the document actually establishes and make uncertainty visible. Where technical language carries a standardised operational meaning, a clear explanation may accompany the source code but should not replace it with an unsupported interpretation. The current issuing authority remains the final source for consequential use.
Verification checklist before release
- Confirm exact issuing authority and product family
- Confirm source advisory number and version
- Confirm UTC date-time groups and their separate functions
- Confirm observation versus forecast labels
- Confirm every coordinate, hemisphere and polygon point
- Confirm vertical units and flight-level notation
- Confirm uncertainty and missing-data labels
- Confirm remarks and next-advisory fields
- Confirm all translated source codes with a second reviewer
- Identify fictional teaching cases and archival examples clearly
- Keep traceable source and retrieval links
- Never substitute the translated article for official flight decisions
Authoritative references and further reading
This guide is educational. The linked services and institutions publish the current technical material and definitions used in real-world practice.
NOAA National Weather Service: Volcanic Ash Advisory Centers
NOAA/NWS: Volcanic Ash and Ashfall
Anchorage VAAC official advisories
ICAO: Volcanic Ash SIGMET issuance, 2025 technical discussion
Contextual next routes
If the difficulty is vocabulary rather than volcanic ash reporting, practise the language distinctions separately. If it is the type of document, open the matching canonical owner instead of trying to make one article answer every translation question.
Top Ways to Translate Correctly — series index
Translate ICAO NOTAM runway closures and validity
Translate METAR and TAF aviation weather
Translate emergency-alert severity and expiry
How English Works: information structure and focus
Final principle: a correct translation lets the reader see what the issuing document actually knows, when it knows it, and what it does not claim. For aviation products, precision is not pedantry. It is how words remain anchored to evidence.