FAQs
Last year we have already released a demo dataset, called TinyEuroCrops, covering areas in Austria, Slovenia and Denmark. This dataset was harmonised with a different, simpler taxonomy (HCATv1).
The dataset and some information about it are available on our website.
We released the data in several ways:
- CSV files (just the labels)
- HDF5 files including the Sentinel-2 reflectance value for a representative pixel (median) of each parcel for one entire year
- GeoJSONs (Geo-referenced parcel polygons including the crop information)
Please have a look at our BIDS22 paper for more details on the dataset.
As we only include data collecting within the subsidy control, we usually have only one type of crop per parcel per year in our dataset. Some countries (e.g. Austria and Portugal) submit multiple crops per field, which can be extracted if you have a closer look at the attribute tables or our country mappings.
We do not have any information about the crop yield. We know, it would be cool, but it's unfortunately not in the data we're given.
You can download our harmonised shapefiles from our Sync&Share platform or in raw format from the county websites which you find in our Wiki. As we provide the mappings from the raw data to our harmonised taxonomy, you can also use these to harmonise data from years which are currently not represented in EuroCrops.
Right now, we only harmonised one year per country, but we will proceed to add more within the next year.
