A Multi-Temporal Analysis of Archaeological Site Destruction using Landsat Satellite Data and Machine Learning, Moche Valley, Peru

The destruction of archaeological sites and the loss landscapes remains a global concern as populations urban areas continue to expand. Archaeological are not only significant local communities, national identities, modern tourist economies but also provide critical knowledge past socio-cultural interactions, settlement patterns, human-environment relationships, risk mitigation strategies. While site have remained outside traditional land use cover change (LULCC) studies, they form agricultural use. By conceptualizing within science, this study provides an innovative approach for assessing “what's left” historically surveyed landscapes. Using Random Forest algorithm Landsat satellite data, quantifies attributed in Peru's lower Moche Valley between 1985-2020. Over 400 previously recorded during Chan Chan-Moche Project (CCMVP, 1969-1974) analyzed. Results indicate that less than quarter original CCMVP remain on landscape. primary drivers LULCC include population growth, migration, government policies, while secondary heritage values. Positioning survey data science integrating machine learning techniques can benefit historic reassessments globally socio-economic conditions underly dynamic landscape changes.