Combining time-series of Sentinel-1 and Sentinel-2 for soil organic carbon estimation and mapping. Application to agricultural soils of a catchment area in Brittany, France - INRAE - Institut national de recherche pour l’agriculture, l’alimentation et l’environnement Access content directly
Conference Poster Year : 2023

Combining time-series of Sentinel-1 and Sentinel-2 for soil organic carbon estimation and mapping. Application to agricultural soils of a catchment area in Brittany, France

Youssef Fouad
Didier Michot
Amaury Bardelle
  • Function : Author
Pascal Pichelin
  • Function : Author
  • PersonId : 972422
Hayfa Zayani

Abstract

The degradation of agricultural soils, in the context of climate change and a growing world population, requires the urgent implementation of sustainable management of soils to improve their health. Therefore, it is important to produce maps that provide accurate information on the current state of soils and allow monitoring of soil properties changes over space and time. This is particularly true in the context of the 4 per 1000 initiative (Arrouays et al., 2019), which requires quantifying soil organic carbon storage in order to assess the relevance of changes in agricultural practices. This study is carried out in the framework of the STEROPES project of the European Joint H2020 Program SOIL1 aiming to update SOC maps based on the use of Sentinel satellite time-series (Vaudour et al., 2022). Our main purpose is to evaluate the accuracy of SOC content estimates predicted using Deep Neural Network (DNN) algorithm and combined time-series of Sentinel-2 (S-2) images and soil moisture derived from Sentinel-1 (S-1) radar images. Our approach was implemented to map SOC content of 12 agricultural fields over the Naizin catchment area (12 km²) in Brittany, western France. In October 2020, 55 composite soil samples were collected from the top 5 cm within these fields and Sentinel time-series were constituted using images acquired between September 2020 and August 2021. Setting the cloud cover threshold to 5% resulted in 24 usable S-2 images. After testing different combinations for the DNN input data, the best results in estimating SOC contents were achieved with time-series combining S-2 images with several spectral indices derived from S-2 bands and soil moisture derived from S-1 images. Finally, our results showed that the implemented approach resulted in a relatively accurate SOC content map. References Arrouays, D., Horn, R., 2019. Soil Carbon - 4 per Mille – an Introduction. Soil Tillage Research 2019, 188, 1-2, doi:10.1016/j.still.2019.02.008. Vaudour, E., Gholizadeh, A., Castaldi, F., Saberioon, M., Borůvka, L., Urbina-Salazar, D., Fouad, Y., Arrouays, D., Richer-De-Forges, A.C., Biney, J., Wetterlind, J., van Wesemael, B., 2022. Satellite Imagery to Map Topsoil Organic Carbon Content over Cultivated Areas: An Overview. Remote Sensing 2022, Vol. 14, Page 2917 14, 2917 https://doi.org/10.3390/RS14122917
No file

Dates and versions

hal-04352914 , version 1 (19-12-2023)

Identifiers

  • HAL Id : hal-04352914 , version 1

Cite

Youssef Fouad, Didier Michot, Amaury Bardelle, Pascal Pichelin, Nicolas Baghdadi, et al.. Combining time-series of Sentinel-1 and Sentinel-2 for soil organic carbon estimation and mapping. Application to agricultural soils of a catchment area in Brittany, France. Soil Mapping for a Sustainable Future. 2nd joint Workshop of the IUSS Working Groups Digital Soil Mapping and Global Soil Map, Feb 2023, Orléans, France. ⟨hal-04352914⟩
48 View
0 Download

Share

Gmail Facebook X LinkedIn More