New geostatistical approach to the analysis of structural lineaments: Case study in the Zat region (Western High Atlas, Morocco)

Abstract

This paper presents a geostatistical study of structural lineaments acquired from satellite data processing (L8: Landsat 8 OLI and AD: ASTER DEM) of the Zat area (High Atlas, Morocco). The lineaments are identified by their metric parameters, by defining three relationships (R1: length-spacing, R2: number-spacing, and R3: number-length) measured with linear regression analysis. The relations R1, R2, and R3 of the L8 set have respective relational weights of β1 (−0.079), β2 (−0.045), and β3 (0.083), giving a relational cascade with a mediating effect (Number → Length → Spacing). According to this model, the number and length grow proportionately, whereas the spacing varies inversely. However, the relations R′1, R′2, and R′3 of the AD set manifested the respective relational weights of β′1 (−0.043), β′2 (0.11), and β′3 (0.037) by materialising a different relational cascade (Number → Spacing → Length). This model shows that the number and spacing vary proportionately, while the length changes inversely. We illustrated that the L8 model matches the tectonic logic of the studied area better than the AD model. As a result, L8 data outperform AD data in structural mapping using remote sensing.