An Ensemble of Geostatistical Simulated Realizations Using a Clustering Algorithm to Reproduce Model-Based Statistics: Case Study of a Gravel Reserve at Kivalina, Alaska

Abstract

This paper demonstrates the convergence of model-based statistics from multiple simulated realizations. Theoretically, the convergence of realization statistics is guaranteed over the number of realizations that are independent among themselves. The rate at which realization-based statistics converges with model-based statistics is important and must be assessed. However, due to poor selection of the random number generator, the generated realization might be far from mutual independence. We use the k-means clustering algorithm to select nearly independent realizations from a set of realization models. We apply the proposed algorithm to a coastal erosion problem in Alaska to estimate the amount of gravel.