Wine Weather Score Explorer
Select a country, wine type, region, and vintage to explore weather-based vintage scores.
Wine regions and weather stations
Weather response curves
How to read the weather score
The score summarizes how favorable the historical weather conditions were for the selected wine region and vintage.
| Predicted weather rating | The algorithm was trained on wine ratings from 1 to 5. This value estimates the average rating expected for wines from the selected region and vintage based on weather conditions alone. |
| Growing season temperature | Average temperature from April to September, representing the main grape growing season. |
| August precipitation | Total precipitation in August. This period is important because rainfall shortly before harvest can influence ripening and disease pressure. |
| Winter precipitation | Total precipitation from October of the previous year to March of the vintage year, representing winter water availability before the growing season. |
What does this score mean?
This website estimates how favorable the weather conditions were for a German wine region and vintage. The selected weather variables are inspired by the Ashenfelter wine quality approach, which used climate data to predict Bordeaux wine quality. Here, a different machine learning method was used for German wines.
| What the model does | It compares region-vintage weather conditions with historical wine ratings and estimates whether the weather was favorable or difficult. |
| What the score can show | In the model, weather conditions can shift the expected average rating by 20%. Under favorable conditions, predicted ratings may move roughly from the mid-3 range toward the mid-4 range. |
| What the score does not show | It does not guarantee that an individual bottle is good. Wine quality also depends on winery skill, grape variety, storage, bottle age, and personal preferences. |
| who benefits from the score | The score helps buyers compare vintages when choosing between promising bottles, selecting a special wine, or considering long-term wine investment. |
Modern wineries can reduce year-to-year variation through techniques such as selective harvesting, controlled fermentation, blending, canopy management, and careful cellar work. For this reason, the score should be understood as a general climate-based vintage index, not as a final judgement of wine quality.
Transparency & Project Information
WeinWetterWelt is an independent data science project that combines historical climate data, machine learning, and community wine ratings to estimate how favorable weather conditions were for German wine vintages. The project is fully transparent and open source.
View the full methodology, source code, data processing workflow, and model development on GitHub:
Acknowledgements
This project would not have been possible without publicly available datasets, scientific publications, and open-source tools.
| X-Wines Dataset |
Thanks to the X-Wines authors and community for providing one of the largest publicly available wine rating datasets, containing more than 21 million ratings.
X-Wines GitHub Repository |
| German Weather Service (DWD) | Historical climate data was provided by the Deutscher Wetterdienst (DWD) through their open climate database and accessed using the rdwd R package. |
| Ashenfelter Wine Research | The climate variables used in this project were inspired by the pioneering work of economist Orley Ashenfelter, who demonstrated that weather data can explain a substantial part of vintage variation in Bordeaux wines. |
| Open Source Software | This project was developed using R, QGIS, Random Forest machine learning models, WordPress, Elementor, and several open-source scientific packages. |
Disclaimer: The weather score is a climate-based vintage indicator and not a guarantee of wine quality. Wine quality depends on many additional factors including vineyard management, grape variety, winemaking techniques, storage conditions, and bottle age.