Related Experiment Video
Updated: Jul 3, 2026
11:37
RGB and Spectral Root Imaging for Plant Phenotyping and Physiological Research: Experimental Setup and Imaging Protocols
Published on: August 8, 2017
16.3K
Remote sensing imagery detects hydromorphic soils hidden under agriculture system
Fellipe A O Mello1, José A M Demattê2, Henrique Bellinaso1,3
1Department of Soil Science, Luiz de Queiroz College of Agriculture, University of São Paulo, Pádua Dias Av., 11, Postal Box 09, Piracicaba, São Paulo, 13416-900, Brazil.
Scientific Reports
|July 5, 2023
Summary
Mapping hydromorphic soils (HS) using remote sensing and random forest (RF) identified 14.5% of Brazil
Area of Science:
- Environmental Science
- Soil Science
- Remote Sensing
Background:
- Agricultural expansion threatens water resources, particularly by impacting hydromorphic soils (HS).
- Environmental regulations exist, but effective detection and protection of HS remain challenging.
- Hydromorphic soils play a crucial role in groundwater recharge and riverine ecosystem health.
Purpose of the Study:
- To develop and apply an advanced remote sensing technique for mapping hydromorphic soils in Brazil.
- To assess the distribution of hydromorphic soils within agricultural areas.
- To provide data supporting public policies for hydromorphic soil conservation.
Main Methods:
- A temporal remote sensing strategy was employed to create a synthetic soil image (SYSI).
- Random forest (RF) classification was utilized to map hydromorphic soils.
- Model performance was validated using cross-validation, achieving an accuracy of 0.92.
Main Results:
- Hydromorphic soils were detected using distinct spectral patterns identifiable by satellite sensors.
- Slope and SYSI were the most significant predictors in the RF model.
- Approximately 14.5% of the 735,953.8 km² study area was identified as hydromorphic soils, predominantly within agricultural lands.
Conclusions:
- The developed remote sensing technique effectively identifies hydromorphic soils under agricultural use.
- Soybean and pasture areas showed higher proportions of HS (up to 14.9%) compared to sugarcane (3%).
- This method can enhance the identification of HS, aiding conservation efforts and informing public policy.
