Previous slide
Next slide

Testing data-driven models for soil erosion mapping in vineyards

Soil erosion is a serious threat today, as it can reduce the productivity of land and threaten sustainable agriculture in the long term. Therefore, the study of the rate of erosion is an area of high research interest.

Researchers from the Institute for Soil Sciences and Eötvös Loránd University investigated whether traditionally applied erosion models can be replaced by machine learning models based on remotely sensed data in vineyards located near to each other, and whether the models generated are transferable between similar areas in close proximity. This new method would allow estimation for areas where the necessary data are not available.

The results show that data-driven spatial models can be successfully applied to estimate soil erosion. However, the transferability of the models even between nearby sample areas requires further investigation.

The paper, published in the journal Land, is available at the link below:
Takáts T, Pásztor L, Árvai M, Albert G, Mészáros J. Testing the Applicability and Transferability of Data-Driven Geospatial Models for Predicting Soil Erosion in Vineyards. Land. 2025; 14(1):163.

© Copyright ATK Minden jog fenntartva. 2023