Abstract
The creation of large-scale deserts is a consequence of soil salinization, a major land degradation phenomenon that also disrupts ecological equilibrium. Saline lands can be evaluated, mapped, and monitored in different world regions using satellite imaging and remote sensing techniques. In order to assess the ability of Landsat-8 and Sentinel-2A imagery for monitoring the spatiotemporal variations of electrical conductivity (EC) and to determine the importance of salinity indices for prediction in two seasons (summer and winter) using the systematic sampling method, from surface depth (0––20 cm), 90 soil samples were taken of marginal lands of Sirjan Playa, southeast of Iran. Satellite images for two seasons were acquired close to the sampling time to obtain the soil salinity indices. Four machine-learning algorithms, namely artificial neural network (ANN), decision tree (DT), random forest (RF), and support vector machine (SVM), were applied to estimate the spatiotemporal changes in soil salinity. Results confirmed that the ANN model developed by the Sentinel-2A image had the highest performance (R2 = 0.77, RMSE = 16.1, NRMSE = 27.1) to predict ECe in the winter season. Furthermore, RF presents the lowest error for prediction ECe during the summer, independent type of satellite data used. Additionally, in virtually all of the analyzed models, the Sentinel-2A data produced lower RMSE and NRMSE values than the Landsat-8 data throughout the two seasons. Results also confirmed that among the soil salinity indices, the Vegetation Soil Salinity Index (VSSI) was identified as the most effective predictor. Overall, this study demonstrates the potential of satellite imaging and machine learning algorithms for monitoring and predicting soil salinity, which can contribute to the sustainable management of marginal lands in arid and semi-arid regions. The findings also highlight the importance of selecting appropriate satellite data and salinity indices for accurate and reliable predictions.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.