Map of tillage changes

  1. Español no disponible
  2. English

The SENSAGRI tillage change product consists of binary maps of surface roughness indicating changes related to tillage practices. Tillage can be detected by Earth Observation (EO) data which detect changes in residue cover and surface roughness. Changes in residue cover can be traced by passive spectral systems, while changes in surface roughness play an essential role in the radar response of agricultural soils. Binary maps of surface roughness changes are derived from the Surface Soil Moisture Prototype (SMOSAR) code. EO data can effectively detect tillage practices on a large scale, which consists of turning over the soil by mechanical movement. The tool implements an algorithm using the synergy of Sentinel-1 and Sentinel-2. Firstly, it classifies agricultural surfaces into vegetated and bare (or sparsely vegetated). Then, it applies a multiscale approach to separate the effect of moisture from that of roughness.

Relevance for monitoring and evaluation of the CAP

This product can be helpful in the monitoring and cross-compliance of agricultural tillage practices. However, the information produced by this tool is also helpful for the evaluation of the policy’s impacts on climate, environmental and natural resources. Agricultural practices like tillage are essential when evaluators assess the effects of policy or measures on soil organic matter or erosion, which are affected by ploughing, tilling and other crop practices. Agricultural practices of ploughing and tilling also affect the quantity and quality of water or biodiversity. In addition, these practices are part of carbon farming, i.e. the set of agricultural methods aimed at sequestering atmospheric carbon into the soil and in crop roots, wood and leaves.

The algorithm has been validated using observations of various tillage practices, such as ploughing and rolling, gathered over the Apulian Tavoliere site in Italy and Castilla y León in Spain. The classification results indicated an Overall Accuracy that ranges from 68% (Castilla y León) to 82% (Apulian Tavoliere). The tool is at the proof-of-concept stage and can be applied to other regions and Member States. Transferability depends on the successful calibration of the algorithm to the new area, and the latter depends on access to in-situ data.

Last modification date: 
09/12/2021