Data sources and licences

A Meteon Microclimate produces modified, aggregated estimates from external Earth observation, weather, soil, terrain and map data. The list below contains the main data sources used in the application and in developing and validating the models, together with their attributions.

External providers do not endorse or certify the Meteon Microclimate service. Unless stated otherwise, the data are not published in unmodified form: we align them in space and time, aggregate several sources, and derive indicators from them with our own physical and machine-learning models.

Weather and climate data

• Open-Meteo: hourly and daily weather forecasts, multi-model uncertainty data, recent weather and some fallback terrain data. Data licence CC BY 4.0; for commercial use we use subscription API access. Source: Open-Meteo.com — terms and licence.
• MET Norway: we use Locationforecast data as an independent forecast source, modified and aggregated with Open-Meteo data. Licence: CC BY 4.0. Source: Norwegian Meteorological Institute; licence and data policy.
• NASA POWER: multi-year daily temperature series for computing heat units and climate comparisons. Source: NASA Prediction Of Worldwide Energy Resources (POWER), NASA Langley Research Center; POWER Data Access Viewer.
• HungaroMet: daily measured precipitation from rain gauges near fields in Hungary — the water balance uses it for rain that has already fallen. The value for a field is a distance-weighted average of the nearest gauges. Database: Meteorological Data Repository, HungaroMet Nonprofit Zrt., averaged over the individual values. Source: odp.met.hu.
• Copernicus Climate Change Service / ERA5: reanalysis data for model development, backtesting and some climate comparisons. Attribution: “Generated using Copernicus Climate Change Service information”. Climate Data Store.

Satellite and land cover data

• Fields of The World (FTW), Taylor Geospatial et al.: 2025 field boundaries detected by the PRUE model from Sentinel-2 imagery at 10 m source resolution. For automatic selection we use simplified outlines and merge parts spanning tiles. These are estimated cultivation boundaries; actual cultivation may differ. Licence: CC BY 4.0. Source: Fields of The World.
• Copernicus Sentinel-2: vegetation indices (NDVI and NDRE). We access the data via the Copernicus Data Space Ecosystem and, as a fallback, via Microsoft Planetary Computer. Attribution: “Contains modified Copernicus Sentinel data 2026”. Copernicus Data Space Ecosystem · Copernicus data access.
• Copernicus Sentinel-1 (radar): a cloud-independent backscatter signal for detecting unlogged irrigation and rain that actually reaches the field. We access the radiometrically terrain-corrected (RTC) scenes via Microsoft Planetary Computer. Attribution: “Contains modified Copernicus Sentinel data 2026”. Sentinel-1 RTC.
• Landsat 8–9 Collection 2: land surface temperature signal. Landsat imagery courtesy of the U.S. Geological Survey; we access the data via Microsoft Planetary Computer. USGS Landsat Data Access · Microsoft Planetary Computer.

Soil and terrain data

• SoilGrids: sand and clay content and soil properties derived from them. SoilGrids © ISRIC — World Soil Information; licence: CC BY 4.0. Reference: Poggio et al., SoilGrids 2.0 (2021). SoilGrids documentation.
• Copernicus DEM GLO-30 and ESA WorldCover (tree-free terrain): from the 30 m Copernicus surface model we mask wooded and built-up areas (ESA WorldCover 10 m) and fill them from the elevation of the surrounding open land; from this we compute slope, runoff and the wetness index (TWI). Copernicus DEM: © DLR e.V. 2010–2014 and © Airbus Defence and Space GmbH 2014–2018, provided under COPERNICUS by the European Union and ESA; all rights reserved. ESA WorldCover: © ESA WorldCover project 2021 / Contains modified Copernicus Sentinel data (2021) processed by ESA WorldCover consortium; licence: CC BY 4.0. Copernicus DEM · ESA WorldCover.
• EU-DEM: European elevation data via OpenTopoData. Produced using Copernicus data and information funded by the European Union; we sample the data and turn it into our own terrain indicators. EU-DEM data description and attribution.
• ASTER GDEM: global fallback terrain source via OpenTopoData. ASTER GDEM is a product of METI and NASA. ASTER data description.
• Copernicus DEM / Open-Meteo elevation: fallback elevation source when the more detailed terrain service is unavailable. We do not use the results as a raw DEM but as part of our own derived terrain indicators.

Map, satellite basemap and place search

• Esri ArcGIS World Imagery: satellite basemap of the field editor and the zone map (with an ArcGIS Location Platform subscription). We serve the tiles through our own server (proxy), and the map shows the attribution required by Esri: “Powered by Esri” and the source acknowledgement of the World Imagery service (Esri, Maxar, Earthstar Geographics, USDA, USGS and the GIS user community). Esri terms of use.
• OpenStreetMap: fallback map layer and address/place search data. © OpenStreetMap contributors; data licence: ODbL. The public Nominatim service is used in accordance with its usage policy vonatkozik.

Atmospheric composition

• Copernicus Atmosphere Monitoring Service (CAMS): aerosol optical depth (e.g. Saharan dust) for field-level correction of solar radiation and evaporation, via the Open-Meteo air quality service. Attribution: “Generated using Copernicus Atmosphere Monitoring Service information 2026”. CAMS.

Live station measurements

• Meteostat (CC BY 4.0): we gently correct the model's current temperature with the latest hourly temperature from nearby official stations. Attribution: “Source: Meteostat and its data providers”. licenc.
• openSenseMap (ODC PDDL): latest temperature from community weather stations for the correction — with strict quality filtering (excluding outliers and sun-exposed boxes, at least three agreeing stations) and with lower weight than official stations. Source: openSenseMap — PDDL licenc.

Model training and validation

• Meteostat (CC BY 4.0): we used historical station observations for offline model training and validation. The licence also permits commercial use with attribution. We do not make the underlying station data available; users only receive the models' derived forecasts and risk signals. Attribution: “Source: Meteostat and its data providers” (with the original meteorological services, e.g. DWD/OMSZ). Source: Meteostat — licenc.
• Teagasc Oak Park research results: published research results used to check the late blight risk logic against the literature. See Cucak et al.: Evaluation of the ‘Irish Rules’ (2019).

About the derived results

A Meteon Microclimate The frost, heat, precipitation, water deficit, crop condition and disease risk values shown by the service are computed estimates, not unmodified copies of the data sources and not on-site measurements. Inaccuracy, delay or temporary outage of external data may also affect the derived results. On-site verification is recommended before taking action.

Detailed information on sending coordinates to external services and the related data processing is provided in the Privacy notice .