PulseTrendingJournal ClubResearchersJournalsExplore
Instagram
HomeTrendingJournal ClubExplore
Synapse
⌘+K
Synapse
March 27, 2026Canadian Journal of Soil Science

Spatial prediction and mapping of soil salinity using machine learning and remote sensing covariates

View Full Paper
Ask AI
Bookmark
Share

Authors

GGGrace Tariro GoweraPSPreston SorensonABAngela Bedard-Haughn

Discussion

Loading...

Member takes

Overview

Evaluates remote sensing models for predicting soil salinity in irrigated agroecosystems, indicating efficient mapping techniques.

Key Points

  • The study aims to evaluate the effectiveness of remote sensing-based models to predict soil salinity in different irrigation conditions.
  • Used Landsat 8 imagery to derive vegetation and salinity indices
  • Incorporated geomorphometric variables from a LiDAR-derived digital elevation model
  • Applied Random Forest and Support Vector Machine to model soil salinity at three depth intervals
  • Assessed model performance using an independent test dataset
  • Support Vector Machine outperformed Random Forest with R2 of 0.77 and RMSE of 0.48
  • Random Forest achieved R2 of 0.66 and RMSE of 0.51
  • Models effectively predicted soil salinity across diverse irrigation conditions

Cite This Study

Gowera et al. (2026) studied this question.

synapsesocial.com/papers/69c620ab15a0a509bde1943ehttps://doi.org/10.1139/cjss-2025-0092
View Full Paper
Ask AI
Bookmark
Share