Soil salinization is a major environmental issue in Egypt's semi-arid region, characterized by low precipitations rate, high temperatures, and high evaporation rates, ultimately lowers agricultural sustainability. Consequently, effective soil salinity observation, evaluation and mapping are vital to mitigate the negative impacts of soil degradation and ensure sustainable land management. Therefore, this study focused on estimating and mapping soil salinity in semi-arid lands (Sohag Governorate, Egypt), using PRISMA (PRecursore IperSpettrale della Missione Applicativa) visible-Near-Infra-Red (vis-NIR) hyperspectral images as an advanced tool that offers significant promise for extracting detailed information about the composition and condition of the observed areas. For this study, seventeen representative soil profiles were collected in more than 150 cm depth from the field and subsequently analyzed in the lab to determine their EC (dS/m) values. Moreover, PRISMA hyperspectral image of the investigated area was acquired, processed, and some soil salinity hyperspectral indices were applied such as; Salinity Index (SI), Brightness Index (BI), Normalized Differential Salinity Index (NDSI), Combined Spectral Response Index (COSRI) and Coloration Index (CI). Among the PRISMA hyperspectral image, six bands were significant (12, 23, 25, 31, 32, and 49) and used for calculating the hyperspectral indices. Soil salinity spatial variability maps of the study area were generated using the relation between the developed hyperspectral indices and the observed soil salinity as EC values (dS/m). The study area demonstrated a relatively consistent level of salinity, with soil EC values ranging from 0.9 to 2.10 dS/m and averaging 1.40 dS/m, this moderate salinity, evident in the generated maps, suggests a relatively stable soil condition. The COSRI and CI hyperspectral salinity indices achieved the highest accuracy in estimating soil salinity (EC) with coefficient off determination (R2) values of 0.68 and 0.64 , respectively; and the root mean square error (RMSE) values were 0.26 and 0.29 dS/m, respectively. Other indices such as; normalized difference vegetation index (NDVI), NDSI, BI, and SI had lower capability for estimating soil salinity over the study area. The PRISMA hyperspectral data is a very potential tool for estimating soil salinity in the studied area, therefore, monitoring and analyzing soil salinity are essential for developing effective land management strategies.