Sentinel-MSI and Landsat-OLI Data Quality Characterization for High Temporal Frequency Monitoring of Soil Salinity Dynamic in an Arid Landscape

Although the Sentinel-MSI and Landsat-OLI are designed to be similar, they have different spectral, spatial and radiometric resolutions. In addition, relative spectral response profiles characterizing the filters responsivities of the both instruments are not identical between the homologous bands. This paper analyse the difference between the reflectance in the homologous spectral bands of MSI and OLI sensors, VNIR and SWIR, for high temporal frequency monitoring of soil salinity dynamic in an arid landscapes. In addition, their conversion in term of Soil Salinity and Sodicity Index (SSSI) and in term of Semi-Empirical Predictive Model (SEPM) for soil salinity mapping were compared. To achieve these, analyses were performed on simulated data and on two pairs of images (MSI and OLI) acquired over the same area in July 2015 and August 2017 with one day difference between each pair. The results obtained demonstrate that the statistical fits between SMI and OLI simulated reflectance over a wide range of soil samples with different salinity degrees reveals an excellent linear relationship (R2 of 0.99) for all bands, as well as for SSSI and SEPM. The Root Mean Square Difference (RMSD) values are null between the NIR and SWIR homologous bands, and are insignificant for the other bands. Moreover, the SSSI show an RMSD of 0.0007 and the SEPM express an excellent RMSD around 0.5 dS.m−1 reflecting a relative error between 0.001 and 0.05 for non-saline and extreme salinity classes, respectively. Likewise, the two used pairs of images exhibited very significant fits (R2 ≥ 0.93) for spectral band reflectance's, as well for SSSI and SEPM, yielding a RMSD values less than 0.029 for bands and less than 0.004 for SSSI. While, for SEPM, the RMSD fluctuate between 0.12 and 2.65 dS.m−1, respectively, of non-saline and extreme salinity classes. Accordingly, we can conclude that the MSI and OLI sensors can be used jointly to monitor accurately the soil salinity and it's dynamic in time and space in arid landscape, provided that rigorous preprocessing issues must be addressed before.