In-situ data for global crop mapping

WorldCereal counts on the global agricultural monitoring community to share high-quality in-situ data to train and validate the custom-developed crop classification algorithms. Discover which in-situ data have been collected or which regions needs further initiatives and maybe you can support the development of an up-to-date cropland and crop type map on a global scale.

Cereal production has a central role in achieving food security and reaching the SDG Zero Hunger target. In order to monitor the progress of this target the proportion of productive agricultural area is a key indicator. At present there is no timely, global view of the total agricultural area, nor of the proportion of the different crop types. WorldCereal is therefore developing an efficient, agile and robust EO based system which can provide this information and hereby support timely global crop monitoring services at field scale.
In order to achieve this, WorldCereal counts on the global agricultural monitoring community to share high-quality in-situ data to train and validate the custom-developed crop classification algorithms. Discover which in-situ data have been collected or which regions needs further initiatives and maybe you can support the development of an up-to-date cropland and crop type map on a global scale.

Gathering in-situ data

Research communities, donor agencies, and governments have been stressing the importance of data sharing, harmonization, and collaborative re-use of data for crop monitoring over the decade. However, data discovery, gathering, managing and harmonizing from different sources is a challenging task. Further, ethical, legal, and consent-related restrictions associated with sharing represent a common dilemma faced by international research projects. The WorldCereal team aims to improve this situation and is therefore: 

  1. Discovering in-situ data across the world
  2. Developing and applying data curation and harmonization
  3. Providing standardized access to a metadata catalogue and the in-situ data itself
  4. Building trust and long-term relationships with the global crop mapping community
  5. Building the first global harmonized in-situ data set for agricultural monitoring

Looking for data from regions in Asia

We are currently gathering as many in-situ data as possible from organizations all over the world for training classification algorithms and validation of our newly produced cropland and crop type maps. Important for us to know is the location of the crop type, the observation date, the growing season and information if the field was irrigated or not. Data should be preferably younger than 2017. We are currently targeting in-situ data sets, but validated crop type maps or irrigation maps can also be of value.

We already collected data from various regions in Europe, North and South America and Africa coming from various sources including  the European parcel registrations (LPIS and SIGPAC), ESA Sen2Agri, the GEOGLAM-JECAM sites, the Radiant Earth ML hub, the CGIAR GARDIAN data repository and data from the NASA Harvest initiative, but unfortunately large regions are not yet covered, specifically in Asia.

We have mapped land cover and crop type descriptions to the WorldCereal legends, which are largely based on FAO’s system. Moreover, we annotated the data sets giving credit to the owner. We briefly describe the data set in terms of name, , DOI, license, data type, objective etc providing specific details on the observation method (think of field survey, road surveys, interpretation of high resolution imagery data), accuracy in terms of timing, spatial resolution and classification and summary statistics on land cover and crop type included, governing water management, years included and data size.

Legends

 

Access improved global cropland and crop type maps

Collecting this information has been a huge effort for the WorldCereal consortium. Although algorithm benchmarking testing showed that the selected algorithms perform well in zones with similar conditions regarding climate and cropping systems, we need to collect more data, specifically covering other climate zones and cropping systems. The additional data will allows us to further improve the training of algorithms and validate the global cropland and crop type map.

We believe there is more useful data out there.  For this the WorldCereal team is counting on the global agricultural monitoring community. Would you like to contribute and profit from improved and up-to-date global cropland and crop type maps in your work? Click here for more information and detailed specifications of the data we search for. Or contact us to discuss future collaboration. 

Supporting WorldCereal has multiple benefits such as:

  • easy access to in-situ data sets
  • first-hand access to products 
  • test and validate your new products and algorithms
  • identify huge data gaps and set up joint in-situ data campaigns
  • co-development/learning
  • joint publications
  • visibility and attribution

Worldcereal has a phased approach; where in the first phase we will be demonstrating our products in the following countries, Argentina, Spain, Ukraine and Tanzania, so data sets for these countries are now a high priority for the project. In the second phase we will produce global maps for which even more data is needed.  So, we are more than happy to join forces to develop the best possible products.