Modeling the Spatio-Temporal Dynamics of Mining Lifecycles: A Markovian Approach Using Global MRDS Data

Abstract This study presents a probabilistic modeling of the temporal evolution of mining site exploitation statuses using the open Mineral Resource Data System (MRDS) provided by the USGS. A first-order homogeneous Markov chain model is proposed to estimate transition probabilities between observed mining statuses (Prospect, Active, Past Producer, Closed). The empirical transition matrix is computed from time-ordered site data, revealing asymmetric and absorbing behaviors. We extend the model with a Hidden Markov Model (HMM) to uncover latent regimes that explain hidden dynamics influencing mining site evolution. This work contributes to sustainable resource planning by providing a reproducible framework for understanding mining lifecycle patterns using open environmental data.