About the live forecast
This site displays modelled forecasts from CAMS. Each page identifies the forecast date and any missing or incomplete coverage.
Our data source
The primary pipeline downloads the CAMS European air quality forecasts dataset directly from ADS, using the ensemble at surface level. It provides hourly concentrations on a 0.1° European grid. The six pollen variables are alder, birch, grass, mugwort, olive and ragweed.
CAMS classifies these pollen forecasts as experimental: they are not among the variables routinely validated against in-situ observations in this product. A forecast is not a measured pollen count at your address.
Official CAMS dataset and documentationFrom hourly data to a daily outlook
Times are converted to Europe/London before grouping into dates. A complete daily value is the maximum hourly forecast concentration for that local day. It is not a daily average, an exposure total or an allergy index.
A day normally contains 24 hourly samples, but the clock changes create 23-hour and 25-hour days. Missing, negative or non-finite values do not become zero. If hours are missing, we label the result as a partial, available maximum and show the coverage. If no valid hours exist, we show “No forecast”.
A midnight UTC model run may not cover the beginning of the current British Summer Time day. That day is marked partial rather than silently filled using a different model run.
How pollen levels are shown
We show concentrations in grains/m³ and descriptive bands, not a numeric health score. The bands use the concentration reference points described for the European Environment Agency’s hourly CAMS pollen viewer: 10 and 100 grains/m³ for alder, birch, olive and mugwort; 3 and 50 for grass and ragweed.
On PollenCount, a positive value below the lower reference is labelled “Low”, from the lower reference to below the upper reference “Moderate”, and at or above the upper reference “High”. Zero is “None predicted”. These are our descriptive labels for the reference intervals, not an official CAMS daily risk rating or individual symptom thresholds. The reference points relate to season and peak-period concentrations; a single daily peak cannot establish the start of a pollen season.
Each type’s daily band describes its highest forecast hourly concentration during a complete local day. It is not a daily average or accumulated exposure. For all six types we show the highest band reached by any type; concentrations are not added together. Every type reaching that band is named when it is moderate or high. Selecting one type shows its own band. Sensitivity adjusts general guidance only, never the concentrations or their classification.
A full-day band requires complete hourly coverage for every selected type, including 23- and 25-hour days. Missing or partial days show available raw values without a full-day band. Only complete days are compared in the outlook text. Small changes below the lower reference are not exaggerated into percentage increases.
Other providers may use different pollen types, sampling periods or classifications. Compare matched concentrations and dates rather than treating labels from different services as interchangeable. We no longer convert these values into a 0–10 index.
Personal guidance and everyday care
The pollen and sensitivity choices are optional and apply only while this page is open. We do not infer allergies, save a health profile or send these choices to our servers. Advice is based on NHS hay fever guidance and AAFA pollen allergy guidance. A low forecast does not rule out symptoms, particularly for sensitive people. No medicine or dose is selected by the forecast.
How a city gets its forecast
A city is associated with its nearest model grid point. Coordinates and distances are checked against a downloaded ADS grid and locked in configuration. The reviewed maximum distance is 7 km; changed grids or unverified mappings block publication.
The grid represents an area, not a street or garden. Nearby cities may share a grid point. Our city outlooks describe the covered locations; they are not nationwide observations.
Updates and unavailable data
The daily refresh starts at midnight in Europe/London, automatically following GMT or BST. The newest CAMS model run is not yet available at midnight, so we use the latest available batch. The fourth calendar day may contain only the first one or two hourly forecasts and is labelled partial. Scheduling delays can postpone publication.
The CAMS model run and the time we retrieved the data are different. The model run is when the upstream forecast was initialised; retrieval is when PollenCount downloaded it. A scheduled delay or manual rerun can retrieve the same model forecast later without making the underlying forecast newer. The forecast archive records both timestamps.
The live refresh starts daily at 00:00 UK time (Europe/London, GMT/BST). Publication happens only after validation and a successful build. A failed or delayed update leaves the previous published version available with its original date. Values are never relabelled as a new forecast.
A stale-data notice appears when the model run is more than 48 hours old. Missing values remain missing, and trend lines do not bridge those gaps.
Historical forecast records
We retain dated forecast editions and their source, model time where supplied, retrieval time and publication status. A retained edition records what was forecast at that time, not a later observation. Later updates do not overwrite earlier editions. Dates before collection began have no record; each city’s “Past forecasts” link opens the verified publication archive. Dates in that browser are forecast-valid dates, and may include forecasts issued before the date occurred. They must not be read as observations.
Comparing days and choosing precautions
The outdoor-planning summary compares complete daily concentration bands for the selected pollen type, or the highest band across available types. All tied lower-band dates are shown together. Zero and low concentrations share one planning tier: tiny changes within that tier do not create a recommended day. Partial dates and dates before today in London are excluded. The summary does not rank hours within a day.
Precautions respond to the selected day's band, pollen type and sensitivity choice. Elevated grass pollen adds guidance on mowing and grassy areas. A low forecast keeps practical everyday options available, without predicting that someone will be symptom-free. These precautions use the NHS and AAFA references below. Out-of-date forecasts stop producing day recommendations.
Weather and air-quality sources
Weather forecasts are provided by Open-Meteo. Air-quality forecasts use Copernicus CAMS data delivered by Open-Meteo, under CC BY 4.0. PollenCount processes hourly values into dated daily summaries; daily peaks are not daily averages.
The current non-commercial launch uses Open-Meteo’s free hosted service. Advertising, subscriptions or other commercial use will not be enabled until the weather and air-quality delivery moves to appropriately licensed access.
Guide references
Attribution and licence
Source: Copernicus Atmosphere Monitoring Service (CAMS).
CAMS hourly model data are processed by PollenCount into local-day maxima. PollenCount is an independent project, not a CAMS service.
Contains modified Copernicus Atmosphere Monitoring Service information 2026. Neither the European Commission nor ECMWF is responsible for any use that may be made of the Copernicus information or data it contains.
Data date: 2026-10-26 · Europe/London
Dataset licence and citation informationCopernicus Atmosphere Monitoring Service (2020): CAMS European air quality forecasts. CAMS Atmosphere Data Store. DOI: 10.24381/a4005cee. Accessed 2026-10-26.
ENSEMBLE data contributors: METEO FRANCE, Ineris, Aarhus University, MET Norway, Jülich IEK, IEP-NRI, KNMI, TNO, SMHI, FMI, ENEA and BSC (2022). CAMS European air quality forecasts, ENSEMBLE data. Accessed 2026-10-26. Licensed under CC BY 4.0.
Information, not medical advice
PollenCount provides environmental forecast information. It does not diagnose allergies, identify individual triggers or recommend treatment.