- Home
- What Does Stale Air-Quality Data Mean?
Direct answer · stale data
What Does Stale Air-Quality Data Mean?
Stale air-quality data means AirAgora is showing an older valid OpenWeather response because a live refresh failed. The original point timestamps remain visible, technical fetch context stays attached to the response, and the result is explicitly labeled stale; it is not rewritten as a current observation. AirAgora treats forecast cache as fresh for 20 minutes and current-endpoint cache as fresh for 10 minutes, and either may be retained for at most two hours after a provider error. After that limit, or without a valid cached response, the product shows unavailable. Use stale data only as time-bounded context, never as proof of present street-level conditions or universal safety.

Stale means older valid data, not fabricated replacement data
A stale result begins with a provider response that was valid when AirAgora fetched it. A later attempt to refresh that data failed, so the application temporarily retained the older response instead of presenting a blank screen immediately. The visible stale label tells the user that the age has crossed the normal fresh-cache window and that the provider did not supply a successful replacement for that request.
The value itself is not recalculated from a scrape, fixture, alternate provider, or guessed trend. Its pollutant concentrations, OpenWeather 1–5 category, valid hours, point timestamps, and technical fetch context remain the facts from the response. That preservation is why stale data can still provide limited context, and why it must never be described as new.
Freshness differs by endpoint, while stale retention stays bounded
AirAgora considers the forecast endpoint cache fresh for 20 minutes and the current endpoint cache fresh for 10 minutes. During those periods, reuse is normal caching rather than a stale state. If a provider refresh fails, a valid retained entry may be served with an explicit stale label for no more than two hours from the original cached response.
The two-hour period is a maximum retention boundary, not a promise that stale data will always be shown. Invalid payloads, missing entries, expired retention, unsupported locations, or other fail-closed conditions can produce unavailable instead. AirAgora does not extend point timestamps or reset the cached age merely because another refresh was attempted.
The original point times and technical fetch context remain distinct
A stale response keeps its original point timestamps visible and its fetch context attached in technical provenance. The current planner does not expose an exact fetch-age field, so the stale warning is the actionable public signal that the normal freshness window has passed. The time of the failed refresh remains operational evidence, not a replacement environmental timestamp. If a browser displays the page later, the view time also stays separate.
The valid hours inside the response also keep their original meaning. A retained forecast point for a future hour remains a forecast for that hour, while a point whose valid time has passed does not become a new observation. Staleness describes the age and refresh state of the dataset, not a conversion from modeled data to measured data.
Stale is not the same as missing or unavailable
Stale data contains an older valid response with known provenance. Missing data means a required field or hour is absent. Unavailable means AirAgora cannot return a valid current or retained result under its guarded rules. Those states should not be merged: an older complete window can be more informative than a gap, while a gap cannot be filled with zero or assumed clean air.
AirAgora evaluates only complete two-hour candidates. If the retained response lacks the required valid component data, it cannot support the normal recommendation even if some fields remain visible. The result should expose the incomplete or unavailable state rather than manufacture a clean-looking period from fragments.
When stale context may still help
A recently stale forecast may help a user remember the last provider-backed pattern, compare whether all available periods were already elevated, or decide that a fresh official check is necessary before committing. It can be especially useful when the alternative is no context at all, provided the visible point time, stale warning, modeled status, and grid limitation remain prominent.
Its usefulness shrinks when the activity is imminent, conditions are changing quickly, a localized event is underway, the stale point no longer covers the relevant hour, or the decision carries high personal or institutional consequences. AirAgora does not calculate a universal cutoff for those situations beyond the technical retention maximum.
A stale grid cannot describe current street or indoor conditions
Even a fresh OpenWeather value is a modeled grid estimate rather than a point observation. A stale value adds an age limitation on top of that spatial boundary. It cannot determine what is happening now beside a road, inside a room, at a monitor, along a route, or under a moving smoke plume. The user should not infer that no change occurred simply because the retained screen stayed the same.
The WHO guidance is one reason AirAgora avoids turning any category into a universal safety claim. With stale data, that boundary becomes even more important: the product can disclose the last known provider-backed forecast, but it cannot give medical clearance, diagnose symptoms, or certify an exact local environment.
Use a stale-result checklist before acting
Check the original point times, local valid hours, stale label, location, OpenWeather 1–5 category, completeness, and main pollutant context. AirAgora does not expose exact fetch age in the current planner, so treat the stale label as a boundary rather than trying to infer an undocumented refresh minute. Ask whether the retained hours still cover the activity; if the answer is unclear, seek fresher evidence rather than reassurance.
During severe conditions or official alerts, use the relevant authority and incident information. If AirAgora reports unavailable after retention expires, that is an honest boundary, not permission to assume the previous category continued. The product deliberately prefers a visible failure state over a silent source swap or a timestamp that pretends old data are new.
Direct answers
Frequently asked questions
How long does AirAgora treat air-quality cache as fresh?
The forecast cache is fresh for 20 minutes and the current endpoint cache for 10 minutes. Reuse inside those windows is normal; a stale label applies only when a valid older entry is retained after a refresh error.
How long can stale air-quality data remain available?
At most two hours under AirAgora’s stale-on-error rule, and only when a valid cached response exists. The product can become unavailable earlier when guarded requirements are not met.
Does stale mean the value was measured recently?
No. AirAgora’s OpenWeather result remains modeled forecast data. Stale adds that the dataset is older and was retained after a failed refresh; it does not turn the value into an observation.
Does refreshing the page reset the stale age?
No. The cache age does not reset, while the original point time and stale warning remain. A failed or repeated refresh attempt does not make the retained data newer.
Is stale data the same as missing data?
No. Stale data is an older valid response with provenance. Missing data lacks a required field or hour and must not be interpreted as zero or clean air.
Can stale data tell me whether it is safe to go outside now?
No. It is older modeled grid context, not a current street-level measurement or personal safety determination. Check fresher official information when timing, urgency, or local conditions matter.
Evidence register
Sources and further reading
The links below identify the primary references used on this page. Selection, summary, and presentation are AirAgora modifications.
Keep exploring
AirAgora provides informational planning guidance, not medical advice. Local conditions can differ from modeled forecasts; follow official guidance during severe events.