Learning from Research

ENVISION cultivated crop type maps

The cultivated crop type maps (CCTMs) are an Earth Observation (EO) crop classification module that exploits satellite data and machine learning techniques to validate the declared crop type by a farmer and compliance with specific environmental requirements.

Output Description

This product from the ENVISION project will deliver several Earth Observation (EO) derived services for cultivated crop type mapping consisting of:

The product will make multiple classifications of the crops throughout the year to ensure the confidence of the classification process. The system will utilise both Sentinel-1 and Sentinel-2 data to ensure high accuracy. Crop type maps will be exported via a shapefile transferred via FTP or HTTP in an automated way.

The service will provide: 

  • Alerts based on significant differences between classification prediction and declarations on a higher level early in the cultivation period;
  •  Alerts based on significant differences in crop type between classification predictions and declarations in the middle of the year.

The service will provide: 

  • Alerts based on the Greening-1 compliance rules for each parcel;
  • Compliance maps via a shapefile will be exported to be transferred via FTP or HTTP in an automated way. 

The product’s output will be: 

  • Crop type maps as a shapefile over the registered parcels grouped in a higher class early in the year. 
  • Crop type maps as a shapefile over the registered parcels in a predefined frequency. 
  • Compliance information as a shapefile over the registered parcels in a predefined frequency. 

The product is tested in Lithuania and Cyprus, where it will produce its output with a maximum frequency of three times per month in Lithuania and three times per growing season in Cyprus

Relevance for monitoring and evaluation of the CAP

The crop type maps to be produced by ENVISION aim at compliance checks because they provide alerts at various time points in the growing season and support smart sampling. The crop types and the compliance checks will be provided in shapefiles, a data storage format for storing the location, shape, and attributes of geographic features.  

Evaluation can re-use the data provided by crop type maps to serve many purposes.  

First, they can be used, together with other data sources and other EO tools, in estimating environmental indicators. For example, an evaluator can estimate irrigation water needs using crop type maps, soil maps, meteorological data, and agronomic information. The estimated irrigation water needs is a proxy for the ‘water use in agriculture’ impact indicator.  

Second, crop type maps can evaluate the effects of agricultural policy measures on environmental indicators. For example, an evaluator can use IACS to get information on beneficiaries and non-beneficiaries of water use measures and compare their potential irrigation water needs.  

Third, crop type maps are data sources that can cross-validate and triangulate information received from other sources. For example, a crop type map can cross-validate information related to policy effects on crop allocation and its consequent impacts on environmental indicators.  

The examples above concern water but can be used for other indicators where prior knowledge of the grown crop is essential. For example, crop type maps can contribute, together with other data, to estimate indicators such as the potential nutrient use, the GHG emissions from managed soils, the soil erosion and soil organic matter, crop diversity, and others that depend on the type of soil cover.    

The tool's adoption requires access to EO, the adaptation and application of the algorithms and their training to recognize the crop types of the region or the Member State. Adopting the tool assumes that the IT infrastructure is adequate and that the evaluator can use the data. In general, when using EO, several conditions may limit their utility and functionality. For this tool, the most critical limitation is the extent of inconclusive parcels, i.e. parcels for which there is no definite crop identification. Inconclusive parcels may be due to specific EO factors such as cloudiness or the prevalence of small parcels and also may be due to difficulties in producing the algorithms to train and forecast crop type. Another difficulty is related to the effort and time needed to link the crop type maps with the LPIS and IACS.  

When fully operational, the ENVISION platform will be open-source. However, access to the tool will be given primarily to the partners and selected customers identified by the project. The toolbox will not be publicly available due to its specificities and the need to have an in-depth exchange of information with potential users before its use. Interested stakeholders and prospective users should contact the project holder.

Relevance of the output per CAP Objectives

  • Specific Objective 4 - Climate change action
  • Specific Objective 5 - Environmental care
  • Specific Objective 6 - Preserve landscape and biodiversity

Additional output information

Data collection systems used:

  • IACS/LPIS 
  • Copernicus

Type of output:

  • New / improved data for M&E
  • Visualisation tools

Associated evaluation approaches:

  • Desk research 
  • Data analysis 
  • Impact evaluation ex post 
  • Impact evaluation ongoing

Spatial scale:

  • Parcel
  • Farm holding

Project information

Envision Logo

The ENVISION project - Monitoring of Environmental Practices for Sustainable Agriculture Supported by Earth Observation - aims to:

  • Address the need for continuous and systematic monitoring of agricultural land, shifting the focus from fragmented monitoring limited to specific fields and dates to territory-wide and all-year-round monitoring.
  • Develop and pilot innovative tools for spatially continuous, uninterrupted large-scale monitoring of farm management activities regarding sustainability and compliance with the CAP's agri-environmental objectives. 
  • Develop a toolbox that will offer a service for monitoring sustainable agricultural practices and tools that Paying Agencies and Certification Bodies can provide to farmers for adhering to environmentally friendly farming practices. 

ENVISION will develop data products related to: 

  • Cultivated crop type maps; 
  • Analytics on vegetation and soil index time series including soil erosion; 
  • Distinctions between organic and conventional farming practices; 
  • Grassland mowing/ploughing;
  • Soil Organic Carbon (SOC) to support the implementation of the SOC business case.

 

Project’s timeframe: 01/09/2020 – 31/08/2023

Contacts of project holder: DRAXIS Environmental SA (info@envision-h2020.eu

WebsiteENVISION: https://envision-h2020.eu/

Territorial coverage: 

Belgium, Cyprus, Greece, Lithuania, Slovenia

Resources