Multisensor and Multiscale Remote Sensing Data Integration Approaches for Archaeological Looting Analysis in Peru

One of the main risk factors which affect the archaeological heritage is the looting which causes the destruction of stratigraphic data and the 'cultural context' that are lost forever and cannot be restored any more. The use of appropriate technologies, including Earth Observation (EO), for the identification of sites exposed to the destructive action of grave robbers, the timely detection and quantification of the extension of the looted areas is a mandatory step for setting effective in situ monitoring strategies in particular for remote and inaccessible archaeological areas. In this paper a remote and close range sensing based approach, including satellite and UAS based image processing, and GPR, has been applied to the Inca settlement of Paredones, in desert environment in Southern Peru, for the detection of looting features. The approach includes also the use of an automatic algorithm named ALFEA (Archaeological Looting Feature Extraction Approach), developed and already experienced by the same authors for other desert sites in South, Central and North of Peru. ALFEA is based on the integration of spatial autocorrelation statistics, unsupervised classification and segmentation. To improve the capability in discriminating and extracting looting features a segmentation step including the setup of multi threshold parameters has been added (ALFEA-I method). Close range analyses based on UAS have been performed with the twofold aims: i) to validate, along with in situ surveys, ALFEA-I method, thus evidencing a good rate of success; ii) to analyze the morphological characteristics of looting features, useful to understand the technique and the depth of digging. With this latter respect complementary information have been provided by also GPR prospecting.