Integration Vis-NIR Spectroscopy and Artificial Intelligence to Predict Some Soil Parameters in Arid Region: A Case Study of Wadi Elkobaneyya, South Egypt

Understanding and determining soil properties is reflected in improving farm management crop production. Soil salinity, pH calcium carbonate are among the factors affecting soil’s physical chemical properties. Hence, their estimation very important for agricultural management, especially arid regions (Wadi Elkobaneyya valley, located northwest of Aswan Governorate, Upper Egypt). The study objectives were to characterize develop prediction models (CaCO3) using integration analysis spectral reflectance vis-NIR spectroscopy. To achieve objectives, three multivariate regression models: Partial Least Squares Regression (PLSR), Multivariate Adaptive Splines (MARS) Square-Support Vector (LS-SVR)); two machine learning algorithms, Random Forest (RF) Artificial Neural Networks (ANN) used. Ninety-six surface samples collected from area at depths 0–5 cm. data divided into a calibration dataset (70% total) validation (30% total dataset). obtained results represent that PLSR model was best parameters where R2 predictability = 0.68 0.52, respectively. LS-SVR predict Electrical Conductivity (EC) Calcium Carbonate content, with 0.70 0.74 0.26 0.47 validation, On other hand, implemented algorithm showed RF CaCO3, as 0.82 0.57 Nevertheless, predicting EC ANN, an 0.96 64 validation. show advantages EC, CaCO3 by Vis-NIR Therefore, spectroscopy considered faster more cost-efficient can be further used environmental monitoring precision farming.